tonsillectomy modern techno

ORIGINAL ARTICLE

Modern Technology–Assistedvs Conventional Tonsillectomy

A Meta-analysis of Randomized Controlled Trials

Vangelis G. Alexiou, MD, MSc; Mary Sheryll Salazar-Salvia, MD, MSc;Paul N. Jervis, BDS, FRCS; Matthew E. Falagas, MD, MSc, DSc

Objective: To systematically review evidence regard-ing modern technology–assisted tonsillectomy pertain-ing to operative time, intraoperative and postoperativebleeding, postoperative pain, and other outcomes.

Design: A systematic search for randomized controlledtrials comparing total tonsillectomies performed usingvessel sealing systems (VSS), Harmonic Scalpel (HS), orradiofrequency ablation (ie, Coblation) with the con-ventional technique of cold steel and/or electrocauterydissection (CS/EC). Estimation of odds ratios and 95%confidence intervals (CIs), weighted mean differences(WMD), or standardized mean difference (SMD), as ap-propriate.

Patients: Thirty-three randomized controlled trials study-ing a total of 3139 patients were included in this meta-analysis.

Main Outcome Measures: Operative time, periopera-tive and postoperative bleeding, and postoperative pain.

Results: For the VSS group compared with the CS/EC

group, operative time was significantly shorter (WMD),−4.09 minutes; 95% CI, −7.43 to −0.75 minutes; 760 pa-tients), perioperative bleeding was significantly less (SMD,−1.67; −2.80 to −0.53; 355 patients), and postoperativebleeding was significantly less (odds ratio, 0.28; 0.13 to 0.61;792 patients). Pain on the first and seventh postoperativedays was significantly less in the VSS group (SMD, −1.73;95% CI, −3.07 to −0.39; 740 patients; and SMD, −1.46; −2.35to −0.57; 684 patients; respectively). For the HS group com-pared with the CS/EC group, the only studied outcome thatdiffered significantly was perioperative bleeding, which wassignificantly less in the HS group (WMD −37.71 mL; 95%CI, −52.98 to −22.43 mL; 535 cases). No difference wasnoted between the Coblation and CS/EC groups for anyof the studied outcomes.

Conclusions: For tonsillectomies, the Coblation and HStechniques do not provide any significant advantage com-pared with CS/EC. Synthesis of the limited and hetero-geneous data regarding VSSs showed a significant ben-efit in all studied outcomes.

Arch Otolaryngol Head Neck Surg. 2011;137(6):558-570

T ONSILLECTOMY IS ONE OF

the oldest and most com-monly performed otorhi-nolaryngologic proce-dures.1 The advent of

antimicrobial therapy2 and the establish-ment of specific surgical indications3 havesubstantially decreased the number of ton-sillectomies. In 2003 and 2004, a total of50 531 patients underwent tonsillectomywithin English National Health Servicetrusts, of whom 49 765 (98.5%) were elec-tive admissions,4 with a significant healthcare expenditure aggravated by the cost ofpostoperative morbidity.5 Modern tonsil-lectomy is a safe procedure. The conven-tional technique is cold steel and/or elec-trocautery dissection (CS/EC). However,postoperative morbidity in terms of pain,bleeding, and return to normal activity anddiet is notable. Thus, a variety of tech-niques and approaches have been testedover the years. Yet, no definite consensus

has been reached regarding the optimaltechnique with the lowest morbidity rates.

Recent advances in surgical instrumen-tal technology have introduced energy-based devices that are able to simultane-ously dissect tissue and seal vessels. Thepatented energy-based vessel sealing sys-tems (VSS) designed for tonsillectomy use

different technologies to provide a simi-lar dissection-ligation effect. The Liga-Sure Vessel Sealing System (LS) (Valley-lab, Boulder, Colorado) is a hemostaticelectrosurgical device that consists of ahandpiece with a ratcheted scissors mecha-nism that grasps and compresses the tis-sue and an electrosurgical generator thatsenses the density of the tissue bundle, au-tomatically adjusting the amount of en-

CME available online atwww.jamaarchivescme.comand questions on page 540

Author Affiliations:Department ofOtorhinolaryngology,Northampton General Hospital,Northampton, England(Drs Alexiou, Salazar-Salvia,and Jervis); Alfa Institute ofBiomedical Sciences(Drs Alexiou and Falagas) andDepartment of Medicine, HenryDunant Hospital (Dr Falagas),Athens, Greece; andDepartment of Medicine, TuftsUniversity School of Medicine,Boston, Massachusetts(Dr Falagas).

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 137 (NO. 6), JUNE 2011 WWW.ARCHOTO.COM558

©2011 American Medical Association. All rights reserved. at , on June 21, 2011 www.archoto.comDownloaded from

http://www.archoto.com

CHARBEL NASSIF

Highlight

ergy to deliver the optimal amount to denature collagenand elastin within the vessel wall and connective tis-sue.6 The Thermal Welding System (TWS) (Starion In-struments Corp, Saratoga, California), instead of usingelectric current, uses heat and pressure to simultane-ously fuse and divide soft tissue and blood vessels. Themanufacturer claims that it produces less heat and mini-mizes collateral tissue damage. It consists of a simple elec-trically resistant heating wire driven by low-voltage di-rect current.7 The third device of this type is the BiClamp(BC) (Erbe Elektromedizin GmbH, Tubingen, Ger-many), which combines bipolar coagulation with high-frequency current modulation automatically regulated anddosed according to tissue impedance. The manufac-turer claims that it results in pulse-and-pause durationthat offers gentle coagulation, avoiding carbonization. Thecoagulation induces vessel wall swelling and sealing ofthe lumen.8

The Harmonic Scalpel (HS) (Ethicon Endo-SurgeryInc, Cincinnati, Ohio), another device that has been usedto safely perform tonsillectomies, uses ultrasonic en-ergy to vibrate its blade at 55 kHz, providing simultane-ous cutting and coagulation of the tissue. This vibrationtransfers energy to the tissue and leads to superficial de-naturation and coagulation of protein by heating the tis-sue. The temperature of the surrounding tissue reaches80°C. The manufacturer claims that this procedure re-sults in precise cutting with minimal thermal damage.9

Finally, radiofrequency ablation (ie, cold ablation) is atechnology that has been used successfully in various sur-gical specialties.10 Bipolar radiofrequency ablation, re-ferred to by the trade name Coblation (ArthroCare Corp,Sunnyvale, California), was designed for head and neck sur-gery; it ablates and coagulates soft tissue by generating afield of ionized sodium molecules. The ionized plasma layerbetween the device tip and the tissue is produced by a radio-frequency current that passes through a medium of nor-mal saline at a temperature of 40°C to 70°C. This processbreaks molecular bonds and produces a melting tissue effect.The manufacturer claims that the lower temperature pro-vides important benefits, such as improved precision cut-ting and minimal thermal damage in tissue.11

Many studies have been published regarding the useof all these surgical techniques and devices. However, thereported results are, in many cases, conflicting and con-troversial. Thus, we sought to systematically review andsynthesize the available evidence regarding modern tech-nology–assisted total tonsillectomy pertaining to opera-tive time, intraoperative and postoperative bleeding, post-operative pain, and other clinical outcomes. We thencompared them with the conventional technique of CS/ECby performing a systematic review and meta-analysis ofrandomized controlled trials (RCTs).

METHODS

DATA SOURCES

This meta-analysis was conducted according to the guidelinesissued by the Quality of Reporting of Meta-analyses confer-ence.12 To identify relevant RCTs, we systematically searchedPubMed for articles dated through July 14, 2010, and the Coch-

rane Central Register of Controlled Trials for the key word ton-sillectomy using only articles published after January 1, 1990.Furthermore, we reviewed the references of the included RCTsto identify additional resources. We did not seek abstracts ofconference proceedings.

STUDY SELECTION

Two of the authors (V.G.A. and M.S.S.-S.) independently per-formed literature searches to locate potentially eligible re-ports. All RCTs comparing total tonsillectomies performed usingVSS (ie, LS, TWS, and BC), HS, or Coblation with tonsillecto-mies performed using the conventional technique of CS/EC andreporting on operative time, intraoperative and postoperativebleeding, postoperative pain, and other clinical outcomes wereconsidered for inclusion in this meta-analysis. The RCTs re-porting on subtotal (ie, intracapsular) tonsillectomies were ex-cluded. Thus, techniques such as microdebrider and laser-assisted tonsillectomy were excluded. Furthermore, we excludedcase series reporting on fewer than 10 patients. Finally, we ex-cluded RCTs pertaining to the use of Argon Plasma Coagula-tion (Erbe Elektromedizin GmbH), a technology that has beenused for tonsillectomy after successfully having been used inendoscopic procedures.13 Argon Plasma Coagulation–assistedtonsillectomy never has been a widely used technique and hasbeen almost abandoned today.

DATA EXTRACTION

Two reviewers (V.G.A. and M.S.S.-S.) independently col-lected from all eligible articles the first author, year of publi-cation, country of origin, age range of study population, num-ber of patients enrolled, tonsillectomy techniques compared,and studied outcomes. Detailed data regarding operative time,intraoperative and postoperative bleeding, postoperative pain,and other clinical outcomes were tabulated. Also, we individu-ally assessed randomization, generation of random numbers,details of the double-blinding procedure, information regard-ing withdrawals, and concealment of allocation to evaluatethe methodologic quality of each RCT according to a modifiedJadad score.14 One point was awarded for the specification ofeach criterion; the maximum score that a study could achievewas 5.

ANALYZED OUTCOMES AND DEFINITIONS

The outcome measures for this meta-analysis were operativetime, intraoperative and postoperative bleeding, and postop-erative pain. Postoperative bleeding was defined as any pri-mary and secondary bleeding reported by the authors of an RCT.

DATA AND STATISTICAL ANALYSES

Statistical analyses were performed using Review Manager (Rev-Man, version 5.0.24, for Linux; Nordic Cochrane Center, Coch-rane Collaboration, Copenhagen, Denmark). The heteroge-neity among the RCTs was assessed with the I2 statistic and a�2 test; P�.10 was defined to note statistical significance in theanalysis of heterogeneity. Publication bias was assessed accord-ing to funnel plot asymmetry. Continuous outcomes were ana-lyzed using weighted mean difference (WMD) or standard-ized mean difference (SMD) if different scales were used tocalculate the same outcome15 and 95% confidence intervals (CIs).The SMD is necessary to standardize the results of the studiesto a uniform scale before they can be combined; it expressesthe size of the intervention effect in each study relative to thestudy variability.15 Pooled odds outcomes of this meta-




analysis were calculated by using a fixed-effects model or theDerSimonian-Laird random-effects model if statistically sig-nificant heterogeneity was noted.

RESULTS

SELECTED RCTs

In Figure 1, we present a flow diagram describing theselection process followed to identify the pool of RCTsincluded in the meta-analysis. The PubMed search yielded3929 potentially relevant articles; the search using theCochrane Central Register of Controlled Trials did notreveal any additional relevant RCTs. An additional 6 ar-ticles were retrieved using the references cited in the re-trieved articles. In total, 33 articles16-48 fulfilled the in-clusion criteria for this meta-analysis.

CHARACTERISTICS OF THE SELECTED STUDIES

In Table 1, we summarize the main characteristics of thestudies included in this meta-analysis: 33 RCTs studyinga total of 3139 patients (median, 68; range, 20-316). Sevenstudies compared tonsillectomies performed using VSS (ie,4, 2, and 1 studies for TWS, LS, and BC, respectively) withtonsillectomies performed using CS/CS. Eleven and 16studies compared conventional tonsillectomy usingCS/EC with tonsillectomies performed using HS andCoblation, respectively. Eleven studies were conductedamong children only and 8 among adults only. Fourteenstudies had a mixed-age population. Five studies ran-domized tonsils instead of patients. Regarding qualityscore, 16 of 33 RCTs achieved a score of 3, 3 studiesachieved a score of 4, and 1 study achieved the maxi-

mum score of 5. The rest of the included studies had alow score of 1 or 2 points.

MAIN OUTCOMES DATA EXTRACTION

In Table 2, we present various clinical outcomes sum-marized by our study. Data regarding operative time, peri-operative and postoperative bleeding, and postopera-tive pain were reported by most of the included studies;thus, a meta-analysis was feasible. Return to normal dietand activity, patient satisfaction, need for analgesics, ton-sillar fossa healing, postoperative nausea and vomiting,need for antibiotics, and other complications, such as he-matoma and taste changes, were reported only by cer-tain studies.

META-ANALYSIS

In Figure 2, we present a meta-analysis comparing out-comes of tonsillectomies performed using VSS with ton-sillectomies performed using the conventional tech-nique (ie, CS/EC). Operative time was significantly shorterin the VSS group (WMD, −4.09 minutes; 95% CI, −7.43to −0.75 minutes; 760 patients). Perioperative bleedingwas noted to be less using VSS compared with the con-ventional technique (SMD, −1.67; 95% CI, −2.80 to −0.53;355 patients). Moreover, VSS yielded significantly lesspostoperative bleeding compared with the conventionaltechnique (odds ratio [OR], 0.28; 95% CI, 0.13 to 0.61;792 patients). Pain on the first and seventh postopera-tive days was significantly less in the VSS group (SMD,−1.73; 95% CI, −3.07 to −0.39; 740 patients; and SMD,−1.46; −2.35 to −0.57; 684 patients, respectively).

In Figure 3, we present the meta-analysis compar-ing outcomes of tonsillectomies performed using HS withtonsillectomies performed using CS/EC. No significantdifference was found between the compared groups re-garding operative time (WMD, −0.10 minutes; 95% CI,−6.26 to 6.05 minutes; 655 cases), postoperative bleed-ing (OR, 0.78; 95% CI, 0.50 to 1.23; 1473 cases), and av-eraged postoperative pain (SMD, −0.38; 95% CI, −1.20to 0.43; 517 cases). However, perioperative bleeding wassignificantly less in the HS group (WMD, −37.71 mL; 95%CI, −52.98 to −22.43 mL; 535 cases).

In Figure 4, we present the meta-analysis compar-ing outcomes of tonsillectomies performed using Cob-lation with tonsillectomies performed using the conven-tional technique. No significant difference was foundbetween the compared groups for any of the studied out-comes: operative time (WMD, −0.35 minutes; 95% CI,−2.84 to 2.13 minutes; 406 cases), perioperative bleed-ing (WMD, −4.22 mL; 95% CI, −14.39 to 5.95 mL; 292cases), postoperative bleeding (OR, 0.99; 95% CI, 0.58to 1.69; 1092 cases), and postoperative pain (SMD, −1.56;95% CI, −3.48 to 0.35; 313 cases).

COMMENT

The main finding of this meta-analysis of RCTs is thatthe Coblation and HS tonsillectomy techniques that havebeen used during the past decade in an attempt to de-

Potentially relevant articles retrievedfrom PubMed and Cochrane

3929

Articles excluded asirrelevant or review

3816

Excluded59Partial tonsillectomy ortonsillotomy

21

Method (ie, observation,review, and nonclinical)

22

Letters13Case report or series3

Excluded27Not randomized20Partial tonsillectomy4Argon plasma coagulationtonsillectomy

3

Articles selected for further evaluationafter first screening of title

113

Articles selected for further evaluationafter first screening of abstract

54

RCTs qualified for inclusion in our reviewplus 6 additional articles retrievedthrough the references of the above;a total of 33 RCTs were included inthe meta-analysis.

27

Figure 1. Flow diagram of the reviewed studies. RCTs indicates randomizedcontrolled trials.




crease postoperative morbidity in terms of pain and bleed-ing do not provide any significant advantage over the con-ventional CS/EC technique. The only outcome thatdiffered significantly in tonsillectomies performed usingHS compared with those using CS/EC was perioperativebleeding. However, it is commonly accepted that this vari-able does not have any measurable clinical significance.Regarding postoperative bleeding and pain in tonsillec-tomies performed using HS, the meta-analysis plot(Figure 3) identified 2 outlier studies19,26 reporting sig-nificant benefits that were not verified by the total ORand SMD, respectively. However, none of the 16 RCTscomparing Coblation with CS/EC showed significant ben-efits regarding postoperative bleeding. This finding wasverified by the meta-analysis, including a total of 1092patients; no benefit of Coblation over CS/EC was ob-served regarding the most important clinical outcomeof postoperative bleeding. Regarding other outcomes,the pooled population was considerably smaller: 406,292, and 303 patients for operative time, perioperativebleeding, and postoperative pain, respectively. Thus,results should be considered cautiously. In all out-comes, HS and Coblation were at least equivalent to theconventional technique. However, equivalence may not

be enough to justify a change in the current clinicalpractice.

The meta-analysis of the RCTs reporting the use ofVSS showed a statistically significant difference in favorof this technique compared with CS/EC for all studiedoutcomes: operative time, perioperative bleeding, and,most important, postoperative bleeding and pain. Of in-terest, none of the included VSS studies was highly pow-ered enough to show statistical significance in the clini-cal outcome of postoperative bleeding. Synthesis of theavailable evidence from 792 patients included in 7 RCTsshowed that the meta-analysis has higher statistical powerto detect an effect on the rare but clinically important out-come of postoperative bleeding. Furthermore, 4 of the 5studies18-22 that evaluated pain in the first postoperativeday showed that it was significantly less for the VSS group.One study17 did not show any significant difference, andanother16 showed that pain was significantly greater inthe VSS group. Combining all the available evidence helpsresolve such controversies and reach a safer conclusionregarding postoperative pain.

This systematic review has included 4 studies ofTWS,16,18,20,21 2 studies of LS,19,22 and 1 study of BC17 thathave not equally addressed all the outcomes included in

Table 1. Main Characteristics of Studies Included in the Systematic Review

Source Country Quality ScoreStudy Population,

Age, yNo. of

PatientsCompared Tonsillectomy

Techniques

Silvola et al,16 2011 Finland 3 16-65 60 VSS (TWS) vs ECLee et al,17 2008 South Korea 3 2-61 215 VSS (BC) vs ECSezen et al,18 2008 Turkey 1 3-28 50 VSS (TWS) vs CSLachanas et al,19 2007 Greece 2 Adults 161 VSS (LS) vs HSStavroulaki et al,20 2007 Greece 3 Adults 32 VSS (TWS) vs CSKaratzias et al,21 2006 Greece 2 17-56 150 VSS (TWS) vs ECLachanas et al,22 2005 Greece 2 16-46 200 VSS (LS) vs CSCushing et al,23 2009 Canada 3 6-47 114 HS vs ECLeaper et al,24 2006 New Zealand 2 6-15 204 HS vs ECParsons et al,25 2006 United States 3 Children 134 HS vs Coblation vs CSKamal et al,26 2006 England 1 3-69 190 HS vs CSOko et al,27 2005 England 3 5-13 122 HS vs CSCollison and Weiner,28 2004 United States 2 Children 28 HS vs CSWillging and Wiatrak,29 2003 United States 3 3-18 120 HS vs ECSheahan et al,30 2004 Ireland 3 16-31 21 HS vs ECWalker and Syed,31 2001 United States 2 1-19 316 HS vs ECAkural et al,32 2001 Finland 3 17-48 28 HS vs ECAksoy et al,33 2010 Turkey 3 �10 50 Coblation vs ECParker et al,34 2009 England 5 4-16 70 Coblation vs CS/ECRoje et al,35 2009 Croatia 3 3-16 72 Coblation vs ECMagdy et al,36 2008 Egypt 4 18-34 40 Coblation vs CS/ECHasan et al,37 2008 Finland 3 18-55 40 Coblation vs ECShapiro and Bhattacharyya,38 2007 United States 3 2-16 47 Coblation vs CSMitic et al,39 2007 Norway 3 4-12 40 Coblation vs ECTan et al,40 2006 Singapore 4 �18 67 Coblation vs ECNoordzij and Affleck,41 2006 United States 2 �18 48 Coblation vs ECPolites et al,42 2006 Australia 2 16-41 20 Coblation vs ECRagab,43 2005 Egypt 3 9-16 200 Coblation vs CSPhilpott et al,44 2005 England 4 Adults 92 Coblation vs CSStoker et al,45 2004 United States 1 3-12 99 Coblation vs ECShah et al,46 2002 United States 1 4-7 34 Coblation vs ECBack et al,47 2001 Finland 3 18-65 37 Coblation vs CSTemple and Timms,48 2001 England 2 4-12 38 Coblation vs EC

Abbreviations: BC, BiClamp; CS, cold steel; EC, electrocautery dissection; HS, Harmonic Scalpel; LS, LigaSure; TWS, Thermal Welding System; VSS, vesselsealing systems.




CHARBEL NASSIF

Highlight

CHARBEL NASSIF

Highlight

Table 2. Data Extracted From the Included Randomized Controlled Trials Regarding Outcomes of the Meta-analysis

Source

OperativeTime, Mean(SD), min

IntraoperativeBleeding, Mean

(SD), mL

No. of PatientsWith

PostoperativeBleeding/TotalNo. of Patients Pain Score, Mean (SD) Other Outcome Information and Scoresa

VSS vs CS/ECSilvola et al,16

201115 (5) vs

13 (5)18 (36) vs 11 (14) 0/31 vs 3/27 POD 1: 2.3 (1.9) vs 1.1 (1.3) (P = .01); MU:

10-point scale;Analgesia: NS differences

In the VSS group, recovery after dischargewas significantly shorter, less pain wasreported, interference with daily activitieswas significantly less, and patientsatisfaction was significantly highercompared with the monopolar group;

Various adverse events after discharge (ie,constipation, tiredness, nausea, andabdominal pain): 26/31 vs 23/27

Lee et al,17

2008Children:

12.0 (3.9)vs14.8 (4.3)(P �.01);

Adults:15.6 (7.0)vs 20.5 vs7.9(P �.01)

Children:1.58 (0.66) vs2.04 (0.75)(P �.01);

Adults: 1.92 (0.83)vs 2.27 (0.87)(P �.05);

Grade of blood loss,scale of I to V:grade I (none),grade II (�20mL), grade III(20-�50 mL),grade IV (50-100mL), and grade V(�100 mL)

Children: 0/45vs 3/47 (P =.24);

Adults: 4/60 vs9/63 (P =.24)

POD 1—child: 0.67 (0.71) vs 1.06 (0.94)(P = .02); adult: 3.78 (1.24) vs 4.10 (1.25)(P = .17);

POD 7—child: 0.47 (0.63) vs 1.13 (1.94)(P = .001); adult: 2.98 (2) vs 5.08 (2.18)

(P �.001);POD 14—child: 0.11 (0.38) vs 0.43 (0.65)

(P = .006); adult: 0.97 (1.34) vs1.79 (1.84) (P = .005); Pain did not differsignificantly on POD 31 in any of thecompared groups;

MU: 4-point visual analog pain scale withpictures of faces for children and 10-pointvisual analog pain scale for adults

Normal activity (in days)— Children:2.93 (1.32) vs 3.83 (2.19) (P = .02);

Adults: 3.52 (1.52) vs 5.83 (2.54) (P �.001);Other perioperative complications (ie, minor

hematoma, anterior pilar tearing,posterior pilar perforation, and tastechange)—Children: 3/45 vs 5/47(P = .71);

Adults: 10/60 vs 21/63 (P = .04);Mean dietary score was significantly higher

in the VSS group in PODs 7 and 14 forchildren and adults

Sezen et al,18

200821.5 (8.81)

vs36.44 (12.07)(P �.01)

17.28 (13.35) vs132.4 (56.9)

0/25 vs 0/25 Pain in VSS group was significantly higherfor PODs 1 and 2 (P �.01);

For PODs 3-7, NS difference;MU: Wong-Baker FACES Pain Scale for

patients younger than 7 years and visualanalog scale for others

Return to normal activity: NS difference;Mean suture number: 0 vs 3 (P �.01);Poor appetite on POD 1: 4% vs 48%

(P �.01)

Lachanaset al,19

2007

15.54 (1.47)vs21 (1.10)(P �.001)

0 vs 73 (20.11)(P �.001)

Primarybleeding:0/50 vs 1/37(P �.001);

Secondarybleeding:1/50 vs 1/37(P �.001)

POD 1: 4.83 (1.3) vs 7.98 (0.67) (P �.001);POD 3: 4.75 (1.79) vs 7.75 (0.79)

(P �.001);POD 5: 4.35 (1.45) vs 6.08 (1.26)

(P �.001);POD 7: 3.33 (1.4) vs 5.33 (1.25) (P �.001);POD 10: 1.33 (1.18) vs 3.00 (0.71)

(P �.001);Overall mean pain score: 3.72 (1.37) vs

6.03 (0.88) (P �.001);MU: 10-point visual analog scale

NA

Stavroulakiet al,20

2007

NA 9.4 (5.20) vs158.44 (30.40)(P �.001)

0/16 vs 3/16 (P= .08)

POD 1: 6.031 (2.41) vs 8.375 (1.58)(P = .003);

POD 2: 5.812 (2.22) vs 7.562 (2.18)(P = .03);

POD 3: 5.593 (2.27) vs 7.312 (1.92)(P = .03);

POD 4: 4.875 (1.70) vs 6.531 (2.40)(P = .03);

POD 7: 3.218 (2.23) vs 4.656 (2.42)(P = .03);

PODs 5-10: NS difference;Cessation of significant pain (pain score of

�7) occurred 3 days earlier in VSS(P = .007);

MU: 10-cm visual analog scale

Analgesic use was higher in CS group;Mean No. of acetaminophens: 4.38 (4.29) vs

6.88 (5.63) (P = .17);Postoperative nausea and vomiting: NS

difference;Tonsillar fossa healing: NS difference

Karatziaset al,21

2006

22.67 (0.38)vs22.23 (0.20)(P � .50)

No measurablebleeding vs 16(range, 0-45 mL)

Primaryhemorrhage:0/81 vs1/69;

Secondaryhemorrhage:1/81 vs 3/69

POD 1: 8.86 (0.15) vs 9.54 (0.07)(P �.001);

POD 3: 8.26 (0.16) vs 9.26 (0.09)(P �.001);

POD 5: 7.90 (0.19) vs 9.12 (0.12)(P �.001);

POD 7: 6.65 (0.23) vs 7.54 (0.18)(P �.001);

POD 10: 2.20 (0.15) vs 4.03 (0.19)(P �.001);

POD 14: 1.04 (0.09) vs 2.51 (0.26)(P �.001);

Overall mean pain score: 5.82 (0.16) vs7.00 (0.15) (P �.001);

MU: 10-point visual analog scale

Mean return to diet (d): 8.44 (0.12) vs12.01 (0.30) (P �.001);

Peritonsillar and uvula edema: 0/81 vs 7/69(resolved in 24 hours with no additionalmedication)

(continued)




Table 2. Data Extracted From the Included Randomized Controlled Trials Regarding Outcomes of the Meta-analysis (continued)

Source



(SD), mL

No. of PatientsWith

PostoperativeBleeding/TotalNo. of Patients Pain Score, Mean (SD) Other Outcome Information and Scoresa

VSS vs CS/ECLachanas

et al,22

2005

15 (1.43) vs21 (1.09)(P �.001)

No measurablebleeding vs 125

Primaryhemorrhage:0/108 vs1/92;

Secondaryhemorrhage:2/108 vs2/92

POD 1: 4.60 (1.66) vs 7.93 (0.64) (P �.001);POD 3: 6.01 (1.95) vs 7.76 (0.73) (P �.001);POD 4: 6.19 (1.43) vs 6.12 (1.27) (P �.001);POD 7: 3.67 (1.88) vs 5.38 (1.25) (P �.001);POD 10: 1.33 (1.18) vs 2.99 (0.70) (P �.001);POD 14: 0 vs 0.33 (0.47);Overall mean pain score: 3.63 (0.91) vs


Peritonsillar edema: 21/108 vs 0/92(resolved in 24 hours with no additionalmedication)

HS vs CS/ECCushing

et al,23

2009

NA NA 1/114 vs 0/114;1 tonsilremovedusing HSand theother usingEC

Averaged pain at rest: 26.4 (28.8) vs27.3 (29.3) (NS difference);

Averaged pain on swallowing: 33.7 (32.3) vs35.2 (32.4);

Pain at rest—POD 10: mean difference, 5.8(P = .04);

PODs 1-9 and 11-14: NS difference;Pain with swallowing—PODs 1, 6, 7, and 8:

mean score differences, 7.2, 8.4, 8.2, and7.8 (P = .02, .002, .03, and .04),respectively;

PODs 2-5 and 9-14: NS difference;MU: 100-mm nonhatched visual analog pain

scale

Delayed postoperative dehydration or poororal intake: none

Lachanaset al,19

2007

14.84 (1.38)vs21 (1.10)(P �.001)

5 (2.77) vs73 (20.11)(P �.001)

Primarybleeding1/43 vs 1/37(P �.001);Secondarybleeding2/43 vs 1/37(P � .001)

POD 1: 4.91 (1.25) vs 7.98 (0.67) (P �.001);POD 3: 4.50 (1.75) vs 7.75 (0.79) (P �.001);POD 5: 4.21 (1.31) vs 6.08 (1.26) (P �.001);POD 7: 3.29 (1.3) vs 5.33 (1.25) (P �.001);POD 10: 1.29 (1.25) vs 3.00 (0.71) (P �.001);Overall mean pain score: 3.64 (1.32) vs


NA

Leaperet al,24

2006

Median(IQR),12 (9-16)vs12 (10.14)(P = .35)

Median (IQR),5 (5-15) vs5 (5-13)(P = .27)

9/103 vs11/101 (P =.60)

Median (IQR)—Child pain score: 4.7 (4.4-4.9)vs 4.2 (4.0-4.4) (P = .002);

Child worst pain: 6.9 (6.7-7.1) vs 6.2 (6.0-6.5)(P �.001);

Child pain with swallowing: 5.9 (5.6-6.1) vs5.2 (5.0-5.5) (P �.001)

Adult pain score: 4.5 (0.1-0.6) vs 4.2 (4.0-4.4)(P = .008);

Adult worst pain: 6.5 (6.3-6.7) vs 6.0 (5.8-6.2)(P = .002);

Adult pain with swallowing: 5.3 (5.1-5.5) vs5.0 (4.8-5.2) (P = .02);

MU: 10-point visual analog pain score

NA

Parsonset al,25

2006

31.5 (9.9) vs21.0 (6.7)(P �.001)

18.2 (24.5) vs11.3 (12.8) (NSdifference)

1/44 vs 2/43(NSdifference)

4.66 (1.67) vs 4.30 (2.10) (NS difference);MU: Wong-Baker FACES Pain Rating Scale

Within 10 PODs, 80.3% of patients achievednormal food intake (P = .08) and91.8% reached normal activity level:NS difference;

Postoperative telephone calls, 3/17 vs 10/19(P = .053)

Kamalet al,26

2006

14.9 (1.94)vs26.16 (1.91)

6.2 (2.54) vs49.38 (3.4)(P = .05)

2/120 vs 12/70 Grades 4-6: 51.5% vs 80%;MU: 1-6 grades

Similar anesthesia requirements;Antibiotic prescription: 12.7% vs 34%

Oko et al,27

200516.2 (4.41)

vs16.7 (3.74)(NSdifference)

3.0 (6.88) vs33.1 (31.26)(P �.001)

8/61 vs 6/61(NSdifference)

Averaged pain score: 1.8 (1.046) vs1.5 (0.994) (P = .003);

POD 1: 2.04 vs 1.69 (P = .047);POD 3: 2.24 vs 1.77 (P = .008);POD 5: 2.09 vs 1.77 (P = .08);POD 7: 1.71 vs 1.33 (P = .08);POD 9: 1.02 vs 0.89 (NS difference);MU: Bieri Faces Pain Scale

Dietary intake scores: POD 1: 0.58 vs 0.4(P �.001);

POD 3: 0.6 vs 0.5 (P = .33);POD 5: 0.62 vs 0.4 (P = .02);POD 7: 0.56 vs 0.25 (P �.001);POD 9: 0.27 vs 0.13 (P = .006);4-point visual analog dietary scale;Readmission rates with pain and dehydration

were 3-fold more common in HS(4.9% vs 1.6%, P = .62)

Collison andWeiner,28

2004

10.9 (1.66)vs7.7 (1.2)(P = .002)

6.2 (4.14) vs58.8 (11.31)(P �.001)

3/28 vs 0/28;1 tonsilremovedusing HSand theother usingCS

3 Hours postoperatively, mean pain score, 3.5vs 4.4 (P = .004);

POD 7: 2.7 vs 2.6 (NS difference);MU: 10-point visual analog scale

NA

(continued)





SourceOperative Time,Mean (SD), min


(SD), mL

No. of PatientsWith

PostoperativeBleeding/TotalNo. of Patients Pain Score, Mean (SD)

Other Outcome Information andScoresa

VSS vs CS/ECWillging and

Wiatrak,29

2003

8.7 (1.29) vs4.55 (0.4)(P �.001)

Only 2 patientsfrom the HS and1 from the ECgroup lost 1 mLof bloodintraoperatively(NS differences)

6/61 vs 3/59(P = .49);

2 patients in HSand 1 in ECgrouprequiredsurgicalintervention

PODs 1, 2, 3, and 14: less pain with HS(P = .04, P = .01, P = .02, and P = .04),respectively;

MU: Wong-Baker FACES Pain Rating Scale

Ability to eat, drink, or swallow andamounts consumed: NS difference;

Pain when speaking or level of ability totalk: NS difference;

Level of daily activity: lower for HS onPOD 1 (P = .01);

Adverse events, dehydration, and fever:NS difference

Sheahan et al,30

2004NA NA Secondary

hemorrhage:1/21 vs 1/21;

1 tonsil removedusing HS andthe otherusing EC

PODs 1, 2, and 7 and week 3: NS difference;MU: 10-point visual analog scale

NA

Walker andSyed,31

2001

NA Minimal bleedingfor both

Secondaryhemorrhage:1/155 vs3/161

NA Return to regular diet within 1 POD:44.3% vs 22.7% (P = .004);

Return to regular diet within 3 PODs:74.2% vs 46.7% (P = .001);

Return to normal activity, POD 1: 27.8%vs 12.0% (P = .01);

Return to normal activity, POD 3: 49.5%vs 22.7% (P = .001);

Readmission for dehydration: 2/155 vs4/161

Akural et al,32

20017 vs 7 0 vs 21 1/14 vs 1/14 Pain score—at rest: at 0-10 hours, 12.3 vs

24.8 (P = .002); first week, NS difference;Second week, 11.5 vs 6.8 (P = .002);With swallowing: at 0-10 hours, 32.5 vs 50.5

(P = .001); first week, NS difference; secondweek, 16.8 vs 9.8 (P = .003);

Day’s worst—first week: NS difference; secondweek: 18 vs 11.75 (P = .002);

Day’s average—first week: NS difference;second week: 9.3 vs 5.5 (P = .004);

Day’s least—2.5 vs 0.5 (P = .01);Otalgia—first week: NS difference; second

week: 10 vs 7 (P = .002);MU: 10-point numerical rating score

Requirement for analgesia;Paracetamol and codeine consumption

persistently high during the firstweek;

From POD 7, codeine consumptiondecreased compared with POD 1;

From POD 8, paracetamol consumptiondecreased

Coblation vs CS/ECAksoy et al,33

20107.3 (1.5) vs

8.1 (1.6)(P = .03)

Minimal bleedingfor both groups

1/25 vs 2/25 3.3 (1.4) vs 3.7 (1.4), NS difference;MU: 10-point visual analog pain scale

PODs 5, 10, and 14: Tonsillar fossahealing scores: NS difference

Parker et al,34

2009NA NA NA (excluded if

postoperativebleedingoccurred)

PODs 1-10: NS differences;MU: Wong-Baker FACES Pain Rating Scale and

Derbyshire Children’s Hospital PaediatricPain Chart

Analgesics requirement: feweranalgesics for Coblation group in thefirst 24 hours

Roje et al,35

2009NA 10.83 (3.41) vs

27.08 (13.22)0/36 vs 0/36 NA Analgesics use, d—mean (range):

4 (0-9) vs 5 (1-8) (P �.05);Return to normal activity (d)—mean

(range): 2 (1-7) vs 4 (1-9) (P �.001);Mean depth in mm of thermal damage

to tonsillar tissue: 428.58 (47.4) vs841.17 (39.7)

Magdy et al,36

2008Coblation vs CS:

median(range),10.5 (7-35) vs14.5 (10-30)(P = .20)

Coblation vs EC:12 (7-40),10 (6-20)(P = .98)

Coblation vs CS:median (range),5 (0-10) vs22.5 (10-75)(P �.001)

Coblation vs EC:median (range),5 (0-10),12.5 (0-25)(P = .21)

Coblation vs CS:0/20 vs 0/20;

Coblation vs EC:0/20 vs 0/20

Coblation vs CS: significantly lower in thecoblation group for PODs 1 and 4-7;

Coblation vs EC: significantly lower in theCoblation group for PODs 1-13;

MU: 10-point visual analog pain scale

Tonsillar fossa healing—Coblation vsCS: NS difference;

Coblation vs EC: POD 7, median (range),faster healing in the Coblation group(P = .004)

Hasan et al,37

2008Median (range),

20.5 (11-45)vs 12 (6-19)

Median: 20 vs 5 1/20 vs 4/20 Daily median pain score: NS difference for allPODs;MU: 10-point visual analog scale

Patient-controlled analgesia devicemedian doses: 10 vs 4 (P = .04)

Shapiro andBhattacharyya,38

2007

5 (0.97) vs7.8 (1.1)(P �.001)

Significantly lowerin the Coblationgroup(P �.001);Graded bloodloss comparison

1/23 vs 0/24 Pain between the compared groups:NS difference

Return to normal activity: NS difference

(continued)





Source



(SD), mL

No. of PatientsWith

PostoperativeBleeding/TotalNo. of Patients Pain Score, Mean (SD)

Other Outcome Informationand Scoresa

VSS vs CS/EC

Mitic et al,39

200725.6 vs 26.6

(NSdifference)

28.25 vs 62.25(significantdifference)

0/20 vs 0/20 6.2 vs 9.6 days with pain score higher than 2(P �.001);


Return to normal activity: 6.6 vs 8.5(P �.001);

Measure used: 5-point visual analog scale;Nutrition score: 6.8 vs 8.9 (P �.001);Measure used: 5-point visual analog scale

Tan et al,40

2006NA NA 2/29 vs 0/38 Daily pain: 3.6 (1.4) vs 3 (1.2) (P = .32);

MU: 10-point visual analog pain scaleReturn to normal activity (d): 7.9 (4.9) vs

10 (6.3) (P = .06);Return to normal diet (d): 11.1 (3.8) vs

12.5 (4) (P = .04);Return to painless swallowing (in days):

NS difference;Use of analgesia tablets per days:

NS difference;Satisfaction score: NS difference

Noordzij andAffleck,41

2006

8.22 (0.6) vs6.33 (0.48)(P = .11)

2.44 (0.58) vs5.39 (0.66)(P = .007)

1/48 vs 3/48; 1tonsil removedusingCoblation andthe other usingEC

Averaged pain score for 14 days: 3.32 (0.1)vs 3.93 (0.1) (P �.001);


Tonsillar fossa healing: NS difference

Politeset al,42

2006

NA NA 1/20 vs 2/20;1 tonsilremoved usingCoblation andother using EC

POD 1: 3.89 (2.60) vs 6.89 (2.05)(P �.001);

POD 2: 4.84 (2.57) vs 6.47 (2.25) (P = .005);POD 3: 5.32 (2.43) vs 6.47 (2.14) (P = .02);For PODs 4-10: NS difference;MU: 10-point visual analog pain scale

NA

Parsonset al,25

2006

28.9 (13.5)vs21.0 (6.7)(P �.001)

21.5 (32.6) vs11.3 (12.8)(P = .15)

1/47 vs 2/43(P = .69)

Mean pain score: 3.27 (1.42) vs 4.30 (2.10)(P = .02);


Marginal difference noted regarding foodintake (P = .08);

No significant difference regarding time toresume normal activity (P = .96);

Postoperative telephone calls, 6/25 vs 10/19(P = .053)

Ragab,43

20058.5 vs 15.5

(P �.001)13 vs 82

(P �.001)1/100 vs 3/100 POD 1: 8.5 vs 9 (P �.05);

MU: 10-point visual analog pain scaleReturn to regular diet, return to normal

activity, and analgesia required:NS difference

Philpottet al,44

2005

NA NA 11/46 vs 8/46 Pain score: Derbyshire Children’s HospitalPaediatric Pain Chart differences for allpostoperative days;

MU: 6-point visual analog pain scale for pain,otalgia, and swallowing

Otalgia, analgesia, and swallowing scores:NS differences for all PODs;Return to normal activity: NS difference;Return to regular diet significantly faster for

the CS group (P = .03)

Stoker et al,45

20047.8 (4.9) vs

8 (2.7)(NSdifference)

98% vs 89% had�15 mL ofbleeding

1/44 vs 1/45 Freedom from analgesics or pain:NS difference;

Patients treated using Coblation tended todiscontinue prescription narcotic usesooner (P = .07);


Return to normal activity: NS difference;Return to normal diet: NS difference;Postoperative calls for mild bleeding, pain,

vomiting, fever, dehydration, no eating, orcoughs: 14/44 vs 23/45 (P = .08);

Posterior and anterior pillar swelling wasmore frequent in ES group (P = .06 and.03)

Shah et al,46

200216.32 (3.2)

vs23.8 (7.9)(P = .002)

83.8 (46.4) vs90.9 (35.3)(NS difference)

1/17 vs 0/17 Pain score: NS differences;MU: Bieri Faces Pain Scale

Return to normal activity and NS difference;Return to normal diet: NS difference

Back et al,47

200127 vs 18

(P �.001)80 vs 20

(P = .02)9/18 vs 8/19

(NS difference)NS differences;MU: 0-100–mm vertical line visual analog

pain scale

Need for analgesia: NS difference;Swelling sensation, difficulty drinking,

eating, opening mouth, and speaking:NS difference

Temple andTimms,48

2001

NA NA 0/18 vs 0/20 POD 1: 4.3 vs 6.4;POD 7: 1.1 vs 6.6;POD 9: 1.0 vs 4.3;PODs 1-9: significant differences (P �.001);MU: 10-point visual analog pain scale

Return to normal diet (d): 2.4 vs 7.6(P �.001)

Abbreviations: CS, cold steel; EC, electrocautery dissection; HS, Harmonic Scalpel; IQR, interquartile range; MU, measures used; NA, not applicable;NS, nonsignificant; POD, postoperative day; VSS, vessel sealing systems.

aMean (SD) except where otherwise indicated.




the meta-analysis. Thus, splitting the VSS RCTs into sub-groups would not permit us to draw any meaningful con-clusions. When further evidence from ongoing or future

studies of these devices is available, it would be useful toperform subanalyses of the different VSS devices, as wellas a comparative analysis. Vessel sealing systems consti-

Operative time

Heterogeneity: tau2 = 18.74; χ2 = 1597.76, df = 6 (P < .001); I 2 = 100%Test for overall effect: z = 2.40 (P = .02)

Study or SubgroupLachanas et al,22 2005Karatzias et al,21 2006Lachanas et al,19 2007

Sezen et al,18 2008Silvola et al,16 2011

Lee et al,17 2008 adults Lee et al,17 2008 pediatric

Total (95% CI)

Weight, %15.415.515.4

10.514.2

14.114.9

Mean Difference(95% CI), min

– 6.00 (– 6.36 to – 5.64)0.44 (0.34 to 0.54)

– 5.46 (– 6.00 to – 4.92)

– 14.90 (– 20.76 to – 9.04)– 2.00 (– 0.58 to 4.58)

– 4.90 (– 7.53 to – 2.27)– 2.80 (– 4.48 to – 1.12)

– 4.09 (– 7.43 to – 0.75)

Mean, min SD, min Total

VSS

15 1.43 9222.67 0.38 8115.54 1.47 50

21.5 8.81 2515 5 31

15.6 7 6012 3.9 45

384

21 1.09 10822.23 0.2 69

21 1.1 37

36.4 12.07 2513 2 27

20.5 7.9 6314.8 4.3 47


Conventional Technique

376

Mean Difference (95% CI), min– 20 – 10 10 20

Favors VSS

FavorsConventionalTechnique

0

Study or Subgroup

Pain on the first postoperative day

Weight, %

Standard Mean Difference (95% CI)– 8 2 6– 6 4– 4 – 2 8

Favors VSSFavorsConventionalTechnique

0

Standard MeanDifference (95% CI)

Lachanas et al,22 2005 14.5Karatzias et al,21 2006 14.1Lachanas et al,19 2007 14.2

Lee et al,17 2008 pediatric 14.4Silvola et al,16 2011 14.3


Stavroulaki et al,20 2007 14.0Lee et al,17 2008 adults 14.5

Total (95% CI)

– 2.56 (– 2.94 to – 2.19)– 5.64 (– 6.36 to – 4.91)– 2.89 (– 3.51 to – 2.28)

– 0.46 (– 0.88 to – 0.05)0.72 (0.18 to 1.25)

– 1.12 (– 1.87 to – 0.37)– 0.26 (– 0.61 to 0.10)

– 1.73 (– 3.07 to – 0.39)

Mean SD

VSS

1.660.151.3

0.711.9

2.41

4.68.864.83

0.622.3

6.0313.78 1.24

Total1088150

4531

1660

391

7.93 0.64 929.54 0.07 697.98 0.67 37

1.06 0.94 451.1 1.3 27

8.375 1.58 164.1 1.25 63

Mean SD Total


349

Study or Subgroup

Pain on the seventh postoperative day

Weight, %

Standard Mean Difference (95% CI)– 6 – 2 4– 4 2 6

Favors VSS


0

Standard MeanDifference (95% CI)Mean SD Total

VSS

Lachanas et al,22 2005 17.2 – 1.05 (– 1.35 to – 0.75)3.67 1.88 108 5.38 1.25 92Karatzias et al,21 2006 16.3 – 4.25 (– 4.83 to – 3.66)6.65 0.23 81 7.54 0.18 69Lachanas et al,19 2007 16.7 – 1.48 (– 1.96 to – 1.00)3.33 1.4 50 5.33 1.25 37

Lee et al,17 2008 pediatric 17.0 – 1.00 (– 1.37 to – 0.62)2.98 2 60 5.08 2.18 63


Stavroulaki et al,20 2007 15.8 – 0.60 (– 1.31 to 0.11)3.218 2.23 16 4.656 2.42 16Lee et al,17 2008 adults 16.9 – 0.45 (– 0.86 to – 0.04)0.47 0.63 45 1.13 1.94 47

Mean SD Total


Total (95% CI) – 1.46 (– 2.35 to – 0.57)360 324

Perioperative bleeding

Standard Mean Difference (95% CI)– 10 – 4– 6 – 2 642 8– 8 10

Favors VSS


0Heterogeneity: tau2 = 1.48; χ2 = 80.12, df = 4 (P < .001); I 2 = 95%Test for overall effect: z = 2.88 (P = .004)

Study or SubgroupStavroulaki et al,20 2007Lee et al,17 2008 pediatricSezen et al,18 2008

Total (95% CI)

Lee et al,17 2008 adults Silvola et al,16 2011

Weight, %14.021.920.422.121.6

Standard MeanDifference (95% CI)– 6.66 (– 8.54 to – 4.79)– 6.66 (– 1.06 to – 0.23)– 2.74 (– 3.53 to – 1.95)

– 1.67 (– 2.80 to – 0.53)

– 0.41 (– 0.77 to – 0.05)0.26 (– 0.26 to 0.78)

Mean, mL SD, mL Total

VSS

9.4 5.2 161.58 0.66 45

17.28 13.35 25

173

1.92 0.83 6018 36 27

158.44 30.4 162.04 0.75 47

132.4 56.9 25

182

2.27 0.87 6311 14 31

Mean, mL SD, mL Total


Study or Subgroup

Postoperative bleeding

Weight, %

Odds Ratio (95% CI)0.002 0.1 10 500

Favors VSS


1

Odds Ratio(95% CI)

Lachanas et al,22 2005 11.4 0.56 (0.09 to 3.42)Karatzias et al,21 2006 17.3 0.09 (0.00 to 1.69)Lachanas et al,19 2007 8.1 0.36 (0.03 to 4.09)

Lee et al,17 2008 adults 12.2 0.14 (0.01 to 2.78)Silvola et al,16 2011 9.4 0.16 (0.01 to 3.53)

Heterogeneity: χ2 = 2.28, df = 6 (P = .89); I 2 = 0%Test for overall effect: z = 3.23 (P = .001)

Lee et al,17 2008 pediatric 29.4 0.43 (0.12 to 1.47)Sezen et al,18 2008 Not estimable

Total (95% CI) 0.28 (0.13 to 0.61)

Stavroulaki et al,20 2007 12.2 0.12 (0.01 to 2.47)

Events

VSS

201

00

40

0

Total1088150

4531

6025

416

16

Events Total


342

32

7 26

90

3

926937

4727

6325

376

16

Total events

Figure 2. Meta-analysis comparing outcomes of tonsillectomies performed using vessel sealing systems (VSS) with tonsillectomies performed using theconventional technique (cold steel and/or electrocautery). CI indicates confidence interval.




tute a new technique; therefore, one should consider manyissues before trying to implement it in everyday clinical prac-tice: cost, training issues, the learning curve for novice sur-geons and additional morbidity associated with the pro-cedure, and, most important, limited evidence to supporta change from the use of CS/EC, the tonsillectomy proce-dure that has proven safe for several decades.

Regarding the quality score, 20 of 33 RCTs achieveda score of 3 or higher, which is generally acceptable forinclusion in a meta-analysis. Only 4 studies18,26,45,46 hada low score of 1 and accounted for less than 12% of theweight of the meta-analysis. This finding may be attrib-uted to the nature of this surgical intervention that does

not permit double-blinding for most of the studied out-comes. Thus, we chose to maintain a lower threshold forinclusion and to include the studies with weak evi-dence. Other methodologic issues regarding the in-cluded RCTs pertain mainly to the data collection meth-ods used (ie, questionnaires, daily diaries, and visualanalog scales) and short follow-up periods. These issuesmay have introduced a degree of bias in our analysis thatneeds to be acknowledged.

The results of this study should be considered in viewof several limitations. The included studies have signifi-cant methodologic heterogeneity, and various scales wereused to measure postoperative pain and perioperative

Study or Subgroup


Weight, %

Odds Ratio (95% CI)0.001 0.1 10 1000

Favors HS


1

Odds Ratio(95% CI)

Akural et al,32 2001 2.2 1.00 (0.06 to 17.75)Walker and Syed,31 2001 6.8 0.34 (0.04 to 3.32)Sheahan et al,30 2004 2.2 1.00 (0.06 to 17.12)

Lachanas et al,19 2007 4.7 1.31 (0.21 to 8.31)Cushing et al,23 2009 1.2 3.03 (0.12 to 75.08)


Collison and Weiner28 2004 1.0 7.82 (0.39 to 158.87)

Total (95% CI) 0.78 (0.50 to 1.23)

Willging and Wiatrak,29 2003 6.4 2.04 (0.48 to 8.55)

Events

HS

Total14

15521

43114

28

764

61

Events Total


131

20

41

03

1416121

37Kamal et al,26 2006 34.9 0.08 (0.02 to 0.38)120 12 70Parsons et al,25 2006 4.6 0.48 (0.04 to 5.46)44 2 43Leaper et al,24 2006 23.7 0.78 (0.31 to 1.98)103 11 101Oko et al,27 2005 12.2 1.38 (0.45 to 4.26)

36

111

31

36

2198 61 6 61

114

28

709

59

Total events

Operative time


Study or SubgroupWillging and Wiatrak,29 2003Collison and Weiner,28 2004Oko et al,27 2005

Lachanas et al,19 2007

Parsons et al,25 2006 Kamal et al,26 2006

Total (95% CI)

Weight, %16.816.816.7

16.8

16.016.8


4.15 (– 3.81 to 4.49) 3.20 (2.44 to 3.96) – 0.50 (– 1.95 to 0.95)

– 6.16 (– 6.70 to – 5.62)

10.50 (6.95 to 14.05)– 11.26 (– 11.83 to – 10.69)

– 0.10 (– 6.26 to 6.05)

SD, min Total

HS

612861

43

44

Mean, min8.7

10.916.2

14.84

31.514.9

1.291.664.41

1.38

9.91.94 120

357

592861

37

43

4.557.7

16.7

21

2126.16

0.41.2

3.74

1.1

6.71.91 70



298


Favors HS


0


Heterogeneity: tau2 = 293.82; χ2 = 226.44, df = 4 (P < .001); I 2 = 98%Test for overall effect: z = 4.84 (P < .001)

Study or SubgroupCollison and Weiner,28 2004Oko et al,27 2005

Lachanas et al,19 2007

Parsons et al,25 2006 Kamal et al,26 2006

Total (95% CI)

Weight, %20.319.6

19.9

19.520.7

Mean Difference(95% CI), mL

– 52.60 (– 57.06 to – 48.14)– 30.10 (– 38.13 to – 22.07)

– 68.00 (– 74.53 to – 61.47)

6.90 (– 1.29 to 15.09)– 43.18 (– 44.10 to – 42.26)

– 37.71 (– 52.98 to – 22.43)

SD, mL Total, mL

HS

2861

43

44

Mean, mL6.2

3

5

18.26.2

4.146.88

2.77

24.52.54 120

296

2861

37

43

58.833.1

73

11.349.38

11.3131.26

20.11

12.83.4 70

Mean, mL SD, mL Total, mL


239

Mean Difference (95% CI), min– 75 – 50 25– 25 50 75

Favors HS


0

Postoperative pain (averaged)


Study or SubgroupOko et al,27 2005Parsons et al,25 2006 Lachanas et al,19 2007Cushing et al,23 2009

Total (95% CI)

Weight, %25.424.923.826.0

Standard MeanDifference (95% CI)0.29 (– 0.06 to 0.65)0.19 (– 0.23 to 0.61)

– 2.08 (– 2.63 to – 1.53)– 0.03 (– 0.29 to – 0.23)

– 0.38 (– 1.20 to 0.43)

Total

HS

614443

Mean1.8

4.663.6426.4

SD1.0461.671.3228.8 114

262

1.54.3

6.0327.3

Mean0.994

2.10.8829.3

SD


614337

114

Total

255

Standard Mean Difference (95% CI)– 10 – 5 5 10

Favors HS


0

Figure 3. Meta-analysis comparing outcomes of tonsillectomies performed using Harmonic Scalpel (HS) with tonsillectomies performed using the conventionaltechnique (cold steel and/or electrocautery). CI indicates confidence interval.




bleeding. To overcome this limitation, the SMD was cal-culated if different scales were used to measure the sameoutcome.14 The SMD is appropriate for depicting signifi-cant differences but does not provide any meaningfulquantification. In general, for continuous outcomes (eg,pain scores, blood loss, and operative time), the notedstatistical heterogeneity was significant. The random-effects model was used to deal with this issue. However,the clinical diversity of the included studies was not strong.

The absence of outliers, especially for the VSS analysis,may indicate a certain degree of publication bias and maybe attributed to the fact that many of the included stud-ies were sponsored by the manufacturers of the relevantdevices. However, we did not note any significant fun-nel plot asymmetry for postoperative bleeding in any ofthe pooled analyses. In all cases, the well-described se-lective publication of RCTs49 and the “file drawer effect”50

also should be acknowledged when interpreting the re-

Operative time


Study or SubgroupStoker et al,45 2004Shah et al,46 2002Parsons et al,25 2006

Aksoy et al,33 2010

Noordzij and Affleck,41 2006Shapiro and Bhattacharyya,38 2007

Total (95% CI)

Weight, %17.812.812.1

18.9

19.319.1


– 0.20 (– 1.85 to 1.45)– 7.48 (– 11.53 to – 3.43)

7.90 (3.55 to 12.25)

– 0.80 (– 1.66 to 0.06)

1.89 (1.67 to 2.21)– 2.80 (– 3.39 to – 2.21)

– 0.35 (– 2.84 to 2.13)

Coblation

Mean, min7.8

16.3228.9

7.3

8.225

SD, min4.93.2

13.5

1.5

0.60.97

Total, min441747

25

4823

204

823.8

21

8.1

6.337.8

Mean, min2.77.96.7

1.6

0.481.1

SD, min


451743

25

4824

Total, min

202


FavorsCoblation


0

Postoperative pain (averaged)


Study or SubgroupTan et al,40 2006Noordzij and Affleck,41 2006 Parsons et al,25 2006Aksoy et al,33 2010

Total (95% CI)

Weight, %25.324.225.425.2

Standard MeanDifference (95% CI)0.46 (– 0.03 to 0.95)

– 6.05 (– 7.01 to – 5.09)– 0.57 (– 1.00 to – 0.15)– 0.28 (– 0.84 to 0.28)

– 1.56 (– 3.48 to 0.35)

Total

Coblation

294847

Mean3.6

3.323.27

3.3

SD1.40.1

1.421.4 25

159

33.934.33.7

Mean1.20.12.11.4

SD


38484325

Total

154

Standard Mean Difference (95% CI)– 8 – 6 – 4 – 2 0 4 86

FavorsCoblation


2

Study or Subgroup


Odds Ratio (95% CI)0.001 0.1 10 1000

FavorsCoblation


1

Odds Ratio(95% CI)

Bäck et al,47 2001Temple and Timms,48 2001Stoker et al,45 2004

Tan et al,40 2006Magdy et al,36 2008


Philpott et al,44 2005

Total (95% CI)

Shah et al,46 2002

Parsons et al,25 2006Noordzij and Affleck,41 2006Polites et al,42 2006Ragab,43 2005

Roje et al,35 2009Aksoy et al,33 2010

Hasan et al,37 2008Mitic et al,39 2007Shapiro and Bhattacharyya,38 2007

Weight, %14.5

3.6

1.5

22.71.7

7.610.97.1

11.1

3.6

14.1

1.7

1.38 (0.38 to 5.03)Not estimable

1.02 (0.06 to 16.89)


1.49 (0.54 to 4.14)

0.99 (0.58 to 1.69)

3.18 (0.12 to 83.76)

0.45 (0.04 to 5.10)0.32 (0.03 to 3.18)0.47 (0.04 to 5.69)0.33 (0.03 to 3.19)

Not estimable1.00 (0.06 to 16.93)


3.27 (0.13 to 84.36)

Events

Coblation

Total181844

2940

46

551

17

474820

31

901

20

111

1111 100

3625

202023

01

101

Events Total


801

00

32

80

192045

382 433 482 203 100

20

01

364 200 200 24

25

46

541

17

Total events



Study or SubgroupShah et al,46 2002Noordzij and Affleck,41 2006Parsons et al,25 2006 Roje et al,35 2009

Total (95% CI)

Weight, %9.633.625.331.5

Mean Difference(95% CI), mL

– 7.10 (– 34.81 to 20.61)– 2.95 (– 3.20 to – 2.70)10.20 (0.13 to 20.27)

– 16.25 (– 20.71 to – 11.79)

– 4.22 (– 14.39 to 5.95)

90.95.3911.3

27.08

Mean, mL35.30.6612.8

13.22

SD, mL


17484336

Total, mL

144

Mean Difference (95% CI), mL– 50 25– 25 50

FavorsCoblation


0

Coblation

Mean, mL83.82.4421.5

10.83

SD, mL46.40.5832.63.41

Total, mL17484736

148

Figure 4. Meta-analysis comparing outcomes of tonsillectomies performed using Coblation with tonsillectomies performed using the conventional technique (coldsteel and/or electrocautery). CI indicates confidence interval.




sults of this meta-analysis. Moreover, the quality scoreof the included RCTs is relatively low, which may be at-tributed to the nature of this surgical intervention thatdoes not permit double-blinding for most of the studiedoutcomes. Finally, as previously mentioned, the VSSanalysis pooled studies of 3 different devices; manufac-turers often claim that their products have unique char-acteristics that make them more efficient than those oftheir competitors. The available evidence that directlycompares different VSS devices in tonsillectomy51 and inother surgical procedures8 shows that these devices are,in fact, similar or equivalent in terms of surgical effec-tiveness and clinical outcomes. Still, the results of thisanalysis should be interpreted with caution.

In conclusion, despite its limitations, this meta-analysis provides evidence that the use of Coblation andHS for tonsillectomy is equivalent to the use of the con-ventional CS/ES technique. Surgeon experience, train-ing, and preferences, as well as cost-effectiveness crite-ria, should be considered. However, statistical synthesisof the limited available data regarding VSS showed a sig-nificant benefit in all studied outcomes. Well-designedand well-performed RCTs are warranted to further in-vestigate the effectiveness of VSS techniques for tonsil-lectomy.

Submitted for Publication: January 9, 2011; final revi-sion received February 23, 2011; accepted March 21, 2011.Correspondence: Vangelis G. Alexiou, MD, MSc,Northampton House, Flat P14, Wellington St, Northamp-ton, England ([email protected]).Author Contributions: All authors had full access to allthe data in the study and take responsibility for the in-tegrity of the data and the accuracy of the data analysis.Study concept and design: Alexiou. Acquisition of data: Alex-iou and Salazar-Salvia. Analysis and interpretation of data:Alexiou, Salazar-Salvia, Jervis, and Falagas. Drafting ofthe manuscript: Alexiou and Salazar-Salvia. Critical revi-sion of the manuscript for important intellectual content:Alexiou, Jervis, and Falagas. Statistical analysis: Alex-iou. Administrative, technical, and material support: Salazar-Salvia. Study supervision: Jervis and Falagas.Financial Disclosure: None reported.Additional Information: Accepted for oral presentationat the First Congress of the Confederation of the Euro-pean Otorhinolaryngology–Head and Neck Surgery; July2-6, 2011; Barcelona, Spain.

REFERENCES

1. McNeill RA. A history of tonsillectomy: two millenia of trauma, hæmorrhage andcontroversy. Ulster Med J. 1960;29(1):59-63.

2. Falagas ME, Vouloumanou EK, Matthaiou DK, Kapaskelis AM, Karageorgopou-los DE. Effectiveness and safety of short-course vs long-course antibiotic therapyfor group A �-hemolytic streptococcal tonsillopharyngitis: a meta-analysis of ran-domized trials. Mayo Clin Proc. 2008;83(8):880-889.

3. American Academy of Otolaryngology–Head and Neck Surgery. Clinical Indica-tors Compendium: Clinical Indicators: Tonsillectomy, Adenoidectomy,Adenotonillectomy. Alexandria, VA: American Academy of Otolaryngology–Head and Neck Surgery; 2000.

4. The Royal College of Surgeons. National Prospective Tonsillectomy Audit. ENTUK—British Association of Otorhinolaryngology Head & Neck Surgery Web site.http://www.entuk.org/members/audits/tonsil/Tonsillectomyauditreport_pdf. Ac-cessed September 25, 2010.

5. van der Meulen J, Browne J. Complication Rates After Tonsillectomy on the Ba-sis of Hospital Episode Statistics. London, England: Clinical Effectiveness Unit,Royal College of Surgeons of England; 2002.

6. Heniford BT, Matthews BD, Sing RF, Backus C, Pratt B, Greene FL. Initial resultswith an electrothermal bipolar vessel sealer. Surg Endosc. 2001;15(8):799-801.

7. Karatzias GT, Lachanas VA, Papouliakos SM, Sandris VG. Tonsillectomy usingthe thermal welding system. ORL J Otorhinolaryngol Relat Spec. 2005;67(4):225-229.

8. Richter S, Kollmar O, Schilling MK, Pistorius GA, Menger MD. Efficacy and qualityof vessel sealing: comparison of a reusable with a disposable device and effects ofclamp surface geometry and structure. Surg Endosc. 2006;20(6):890-894.

9. Shinhar S, Scotch BM, Belenky W, Madgy D, Haupert M. Harmonic scalpel ton-sillectomy versus hot electrocautery and cold dissection: an objective comparison.Ear Nose Throat J. 2004;83(10):712-715.

10. Calkins H. Cooled ablation. J Cardiovasc Electrophysiol. 2004;15(10)(suppl):S12-S17.

11. Woloszko J, Kwende MM, Stalder KR. Coblation in otolaryngology. Proc SPIE.2003;4949:341-352.

12. Farin G, Grund KE. Technology of argon plasma coagulation with particular regardto endoscopic applications. Endosc Surg Allied Technol. 1994;2(1):71-77.

13. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the qual-ity of reports of meta-analyses of randomised controlled trials: the QUOROMstatement. Lancet. 1999;354(9193):1896-1900.

14. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of ran-domized clinical trials: is blinding necessary? Control Clin Trials. 1996;17(1):1-12.

15. Beurs E, Bouter LM. Is the standardised mean difference a suitable measure oftreatment effect? In: Proceedings of the Seventh Cochrane Colloquium; October5-9, 1999; Università San Tommaso D’Aquino, Rome, Italy.

16. Silvola J, Salonen A, Nieminen J, Kokki H. Tissue welding tonsillectomy pro-vides an enhanced recovery compared to that after monopolar electrocautery tech-nique in adults: a prospective randomized clinical trial. Eur Arch Otorhinolaryngol.2011;268(2):255-260.

17. Lee SW, Jeon SS, Lee JD, Lee JY, Kim SC, Koh YW. A comparison of postop-erative pain and complications in tonsillectomy using BiClamp forceps and elec-trocautery tonsillectomy. Otolaryngol Head Neck Surg. 2008;139(2):228-234.

18. Sezen OS, Kaytanci H, Kubilay U, Coskuner T, Unver S. Comparison between ton-sillectomy with thermal welding and the conventional ‘cold’ tonsillectomy technique.ANZ J Surg. 2008;78(11):1014-1018.

19. Lachanas VA, Hajiioannou JK, Karatzias GT, Filios D, Koutsias S, Mourgelas C.Comparison of LigaSure vessel sealing system, harmonic scalpel, and cold knifetonsillectomy. Otolaryngol Head Neck Surg. 2007;137(3):385-389.

20. Stavroulaki P, Skoulakis C, Theos E, Kokalis N, Valagianis D. Thermal weldingversus cold dissection tonsillectomy: a prospective, randomized, single-blind studyin adult patients. Ann Otol Rhinol Laryngol. 2007;116(8):565-570.

21. Karatzias GT, Lachanas VA, Sandris VG. Thermal welding versus bipolar tonsil-lectomy: a comparative study. Otolaryngol Head Neck Surg. 2006;134(6):975-978.

22. Lachanas VA, Prokopakis EP, Bourolias CA, et al. Ligasure versus cold knifetonsillectomy. Laryngoscope. 2005;115(9):1591-1594.

23. Cushing SL, Smith O, Chiodo A, Elmasri W, Munro-Peck P. Evaluating postop-erative pain in monopolar cautery versus harmonic scalpel tonsillectomy. Oto-laryngol Head Neck Surg. 2009;141(6):710-715.

24. Leaper M, Mahadevan M, Vokes D, Sandow D, Anderson BJ, West T. A prospec-tive randomised single blinded study comparing harmonic scalpel tonsillec-tomy with bipolar tonsillectomy. Int J Pediatr Otorhinolaryngol. 2006;70(8):1389-1396.

25. Parsons SP, Cordes SR, Comer B. Comparison of posttonsillectomy pain usingthe ultrasonic scalpel, coblator, and electrocautery. Otolaryngol Head Neck Surg.2006;134(1):106-113.

26. Kamal SA, Basu S, Kapoor L, Kulandaivelu G, Talpalikar S, Papasthatis D. Har-monic scalpel tonsillectomy: a prospective study. Eur Arch Otorhinolaryngol. 2006;263(5):449-454.

27. Oko MO, Ganly I, Loughran S, Clement WA, Young D, Geddes NK. A prospectiverandomized single-blind trial comparing ultrasonic scalpel tonsillectomy with ton-sillectomy by blunt dissection in a pediatric age group. Otolaryngol Head NeckSurg. 2005;133(4):579-584.

28. Collison PJ, Weiner R. Harmonic scalpel versus conventional tonsillectomy: adouble-blind clinical trial. Ear Nose Throat J. 2004;83(10):707-710.

29. Willging JP, Wiatrak BJ. Harmonic scalpel tonsillectomy in children: a random-ized prospective study. Otolaryngol Head Neck Surg. 2003;128(3):318-325.

30. Sheahan P, Miller I, Colreavy M, Sheahan JN, McShane D, Curran A. The ultra-sonically activated scalpel versus bipolar diathermy for tonsillectomy: a pro-spective, randomized trial. Clin Otolaryngol Allied Sci. 2004;29(5):530-534.




CHARBEL NASSIF

Oval

31. Walker RA, Syed ZA. Harmonic scalpel tonsillectomy versus electrocautery ton-sillectomy: a comparative pilot study. Otolaryngol Head Neck Surg. 2001;125(5):449-455.

32. Akural EI, Koivunen PT, Teppo H, Alahuhta SM, Lopponen HJ. Post-tonsillectomypain: a prospective, randomised and double-blinded study to compare an ultrasoni-cally activated scalpel technique with the blunt dissection technique. Anaesthesia.2001;56(11):1045-1050.

33. Aksoy F, Ozturan O, Veyseller B, Yildirim YS, Demirhan H. Comparison of radio-frequency and monopolar electrocautery tonsillectomy. J Laryngol Otol. 2010;124(2):180-184.

34. Parker D, Howe L, Unsworth V, Hilliam R. A randomised controlled trial to com-pare postoperative pain in children undergoing tonsillectomy using cold steeldissection with bipolar haemostasis versus coblation technique. Clin Otolaryngol.2009;34(3):225-231.

35. Roje Z, Racic G, Dogas Z, Pisac VP, Timms M. Postoperative morbidity and his-topathologic characteristics of tonsillar tissue following coblation tonsillectomyin children: a prospective randomized single-blind study. Coll Antropol. 2009;33(1):293-298.

36. Magdy EA, Elwany S, el-Daly AS, Abdel-Hadi M, Morshedy MA. Coblation ton-sillectomy: a prospective, double-blind, randomised, clinical and histopathologi-cal comparison with dissection-ligation, monopolar electrocautery and lasertonsillectomies. J Laryngol Otol. 2008;122(3):282-290.

37. Hasan H, Raitiola H, Chrapek W, Pukander J. Randomized study comparing post-operative pain between coblation and bipolar scissor tonsillectomy. Eur ArchOtorhinolaryngol. 2008;265(7):817-820.

38. Shapiro NL, Bhattacharyya N. Cold dissection versus coblation-assisted ad-enotonsillectomy in children. Laryngoscope. 2007;117(3):406-410.

39. Mitic S, Tvinnereim M, Lie E, Caltyte BJ. A pilot randomized controlled trial ofcoblation tonsillectomy versus dissection tonsillectomy with bipolar diathermyhaemostasis. Clin Otolaryngol. 2007;32(4):261-267.

40. Tan AK, Hsu PP, Eng SP, et al. Coblation vs electrocautery tonsillectomy: post-

operative recovery in adults. Otolaryngol Head Neck Surg. 2006;135(5):699-703.

41. Noordzij JP, Affleck BD. Coblation versus unipolar electrocautery tonsillectomy:a prospective, randomized, single-blind study in adult patients. Laryngoscope.2006;116(8):1303-1309.

42. Polites N, Joniau S, Wabnitz D, et al. Postoperative pain following coblation ton-sillectomy: randomized clinical trial. ANZ J Surg. 2006;76(4):226-229.

43. Ragab SM. Bipolar radiofrequency dissection tonsillectomy: a prospective ran-domized trial. Otolaryngol Head Neck Surg. 2005;133(6):961-965.

44. Philpott CM, Wild DC, Mehta D, Daniel M, Banerjee AR. A double-blinded ran-domized controlled trial of coblation versus conventional dissection tonsillec-tomy on post-operative symptoms. Clin Otolaryngol. 2005;30(2):143-148.

45. Stoker KE, Don DM, Kang DR, Haupert MS, Magit A, Madgy DN. Pediatric totaltonsillectomy using coblation compared to conventional electrosurgery: a pro-spective, controlled single-blind study. Otolaryngol Head Neck Surg. 2004;130(6):666-675.

46. Shah UK, Galinkin J, Chiavacci R, Briggs M. Tonsillectomy by means of plasma-mediated ablation: prospective, randomized, blinded comparison with monopo-lar electrosurgery. Arch Otolaryngol Head Neck Surg. 2002;128(6):672-676.

47. Back L, Paloheimo M, Ylikoski J. Traditional tonsillectomy compared with bipo-lar radiofrequency thermal ablation tonsillectomy in adults: a pilot study. ArchOtolaryngol Head Neck Surg. 2001;127(9):1106-1112.

48. Temple RH, Timms MS. Paediatric coblation tonsillectomy. Int J PediatrOtorhinolaryngol. 2001;61(3):195-198.

49. Dickersin K, Chan S, Chalmers TC, Sacks HS, Smith H Jr. Publication bias andclinical trials. Control Clin Trials. 1987;8(4):343-353.

50. Rosenthal R. The file drawer problem and tolerance for null results. Psychol Bull.1979;86:638-641.

51. Karatzanis A, Bourolias C, Prokopakis E, Panagiotaki I, Velegrakis G. Thermalwelding technology vs ligasure tonsillectomy: a comparative study. Am JOtolaryngol. 2008;29(4):238-241.




tonsillectomy modern techno

Documents