b u r n s 3 4 ( 2 0 0 8 ) 9 2 9 – 9 3 4

Patient controlled sedation using a standard protocol fordressing changes in burns: Patients’ preference, proceduraldetails and a preliminary safety evaluation§

Andreas Nilsson a,*, Ingrid Steinvall b, Zoltan Bak b,c, Folke Sjoberg b,c,d

aDepartment of Anesthesiology and Intensive Care, Division of Perioperative Medicine, Linkoping University Hospital,

581 85 Linkoping, SwedenbThe Burn Unit, Department of Hand and Plastic Surgery, Linkoping University Hospital, 581 85 Linkoping, SwedencDepartment of Intensive Care, Linkoping University Hospital, 581 85 Linkoping, Swedend Faculty of Health Sciences, Department of Biomedicine and Surgery, Linkoping University Hospital, 581 85 Linkoping, Sweden

a r t i c l e i n f o

Article history:

Accepted 10 April 2008

Keywords:

Alfentanil

Anesthesia

Burns

Propofol

Patient controlled sedation

Sedation

a b s t r a c t

Background: Patient controlled sedation (PCS) enables patients to titrate doses of drugs by

themselves during different procedures involving pain or discomfort.

Methods: We studied it in a prospective crossover design using a fixed protocol without

lockout time to examine it as an alternative method of sedation for changing dressings in

burned patients. Eleven patients with >10% total burn surface area (TBSA) had their

dressings changed, starting with sedation by an anaesthetist (ACS). The second dressing

change was done with PCS (propofol/alfentanil) and the third time the patients had to

choose ACS or PCS. During the procedures, data on cardiopulmonary variables, sedation

(bispectral index), pain intensity (VAS), procedural details, doses of drugs, and patients’

preferences were collected to compare the two sedation techniques.

Results: The study data indicated that wound care in burned patients is feasible with a

standardized PCS protocol. The patients preferred PCS to ACS on the basis of self-control,

and because they had less discomfort during the recovery period. Wound care was also

considered adequate by the staff during PCS. No respiratory (respiratory rate/transcuta-

neous PCO2) or cardiovascular (heart rate/blood pressure) adverse events were recorded at

any time during any of the PCS procedures. The doses of propofol and alfentanil and BIS

index decrease were less during PCS than ACS. Procedural pain was higher during PCS but

lower after the procedure.

Conclusion: We suggest that PCS using a standard protocol is an interesting alternative to

anaesthetist-provided sedation during dressing changes. It seems effective, saves

resources, is safe, and at same time is preferred by the patients. The strength of these

conclusions is, however, hampered by the small size of this investigation and therefore

further studies are warranted.

# 2008 Elsevier Ltd and ISBI. All rights reserved.

avai lab le at www.sc iencedi rec t .com

journal homepage: www.e lsev ier .com/ locate /burns

§ The study was performed at the Burn Unit, Department of Hand and Plastic Surgery, Linkoping University Hospital, OstergotlandCounty Council.

* Corresponding author. Tel.: +46 13 22 1834; fax: +46 13 22 2836.E-mail address: andreas.nilsson@lio.se (A. Nilsson).

0305-4179/$34.00 # 2008 Elsevier Ltd and ISBI. All rights reserved.doi:10.1016/j.burns.2008.04.002

b u r n s 3 4 ( 2 0 0 8 ) 9 2 9 – 9 3 4930

1. Introduction

Patients with severe burns have severe pain and anxiety,

which have both psychological and physical effects for the

recovery process [1]. Pain for these patients is elicited not

only from the trauma event, but is also the result of

repeated procedures, dressing changes and physiotherapy.

These procedures are repeated, and often lead to appreci-

able pain, and they emphasise the need for advanced

strategies for the treatment of pain [2]. Morphine, fentanyl,

and alfentanil are commonly used opioids that provide

continuous, analgesia during and after the operation and

the intravenous route is often preferred for rapid pain

control. Self-titration of morphine (patient controlled

analgesia; PCA) has been suggested as an important

technique to meet the increasing and high requirement

for opioids [3]. Successful pain management requires that

pain is regularly assessed and its intensity evaluated as, for

example by a visual analogical scale (VAS), and adequate

pain relief must be provided accordingly [4]. Wound care

and dressing changes are associated not only with pain, but

with factors that are also known to affect the experience of

pain such as anxiety and loss of self-control. The ther-

apeutic challenge is to find methods that control pain and

relieve anxiety and at the same time avoid over-sedation

and a delayed recovery [1].

Patient controlled sedation (PCS) has received increasing

interest when it has been used to improve the conditions of

patients having painful or unpleasant procedures such as

lithotripsy, colonoscopy, or dental procedures [5–7]. It is

important also to others that the patient controlled sedation

technique has been used in these settings (lithotripsy,

dental, colonoscopy) without anesthesia-trained personnel.

Presently, in clinical practice at our hospital, lithotripsy is

done using a propofol-based PCS technique without

anesthesia-trained personnel. The procedure is based on

European Guidelines for sedation and/or analgesia by non-

anaesthesiology doctors [8]. In patients with burns we are

aware of only one dose-finding study that attempted to use

it as an alternative to traditional ways of giving analgesics

and sedatives [9].

The present study was designed to compare PCS, based on a

standard protocol, with routine sedation provided by anaes-

thetists, to assess the feasibility of PCS for adequate dressing

changes, and also to assess patients’ preferences. Close

surveillance of respiratory (respiratory rate/transcutaneously

assessed PCO2) and cardiovascular (heart rate/mean arterial

pressure) data were measured as an estimate of safety, as the

aim is to, in the future and after further refinement, to

implement this strategy in the care of patients with burns

without using specifically trained anaesthesia personnel.

2. Patients and methods

2.1. Study overview

After approval from the local ethics committee and informed

consent of the patients, 11 patients with ASA I or II scores with

burns exceeding 10% total burn surface area (TBSA) were

enrolled in a prospective exploratory study at the National

Burn Centre at Linkoping University Hospital, Linkoping,

Sweden.

Inclusion criteria were: need for analgesia and sedation for

at least two consecutive dressing changes. Patients with

injures of more than 10% or more than 5% full thickness was

asked to participate. Exclusion criteria were ASA III–V, burned

hands (because of difficulties in using the PCS device) or

difficulties in communication or understanding of the proce-

dure. Four patients entered the study but did not fulfil the

protocol because only one or two dressing changes were

needed that required analgesia and sedation.

The study was designed as a single-centre, single-case,

crossover, controlled study. Patients were their own control

group, because they started with dressing changes under

sedation by an anaesthetist (ACS, FS or ZB; anesthesiologists

with 10 years experience of burn care) using routine sedating

techniques. While the patient breathed oxygen and air,

sedation and analgesia were accomplished with intermittent,

intravenous propofol (Propofol-1Lipuro 10 mg/ml, Braun) and

fentanyl (Fentanyl1 0.05 mg/ml, Braun). The second dressing

change was done using PCS as described below. At the third

dressing change the patients were asked to choose one of the

two techniques.

During the sedation procedures a protocol was used to

collect cardiopulmonary data (heart rate, non-invasive blood

pressure, saturation, and respiratory rate). Transcutaneous

PCO2 were collected using TCM3—TINA (Radiometer, Copen-

hagen, Denmark). Intensity of pain was assessed using a 11-

point visual analogue scale (VAS) when patients were able to

answer during the procedure, after finishing the dressing

changes, and 10 min later. Complications were recorded,

whether it was possible to treat the wounds adequately (yes

or no), and the duration of treatment. The bispectral (BIS)

index was monitored using the A-2000 BISTM system (Aspect

Medical Systems, Natick, MA, USA). The recording of all these

variables started before the onset of sedation and every 3 min

during the procedure, and up to 30 min afterwards. If a lower

BIS index appeared on the display between measurement

points, the lowest observed index was recorded. Within 2 h of

completion of the dressing change, each patient was asked

about how they experienced the sedation and the awakening

from it.

2.2. Patient controlled sedation

One hour before the dressing change all patients were given

their regular daytime analgesics. These comprised: acetami-

nophen (Panodil1 1 g, GlaxoSmithKline Healthcare) and a

long-acting opioid, Oxycodone (Oxycontin1 between 5 and

40 mg). Before the start of the procedure or the anticipated

pain the patients were asked to give sedation or analgesia by

the PCS device whenever they felt pain or anxiety. Anticipated

painful events were predicted during the procedure and

conveyed to the patients so that they could prepare them-

selves by giving analgesia and sedation.

Propofol, 20 mg/ml (Propofol-1Lipuro 20 mg/ml, Braun,

Sweden) and alfentanil 0.5 mg/ml (Rapifen1 0.05 mg/ml,

Janssen-Cilag, Sweden) were mixed to 14.8 mg/ml of propofol

and 0.13 mg/ml of alfentanil in the final solution. A bag

Table 1 – Details of the patients

Patient TBSA Age ACS PCS

Time Propofol Morphine Time Propofol Morphine

1 15 74 64 325 10 74 153 10

2 35 32 69 500 40 85 282 18

3 12 51 91 480 20 52 140 9

4 12 82 33 170 10 50 89 5

5 15 67 81 234 10 62 195 12

6 18 23 130 632 25 60 271 17

7 16 77 82 320 15 113 191 12

8 10 68 65 395 15 50 164 10

9 8 62 44 412 15 56 102 6

10 19 59 56 490 20 78 173 11

11 20 38 47 385 25 63 369 23

Mean 16.4 57.5 69.3* 394.8** 18.6*** 67.5* 193.5** 12.1***

S.D. 7.2 19.3 26.7 130.7 9.0 19.0 83.5 5.3

ACS = anaesthetist controlled sedation; PCS = patient controlled sedation.

The doses of propofol and alfentanil were given during ACS and the first PCS.

Values as mean and standard deviation (S.D.). Doses in milligrams and time in min. Total burned surface area (TBSA).

*P = 0.859 (NS); **P = 0.003; ***P = 0.007. Statistical differences are shown between propofol and morphine.

b u r n s 3 4 ( 2 0 0 8 ) 9 2 9 – 9 3 4 931

containing the mixture was connected to an electromecha-

nical pump (Graseby 9300 PCS, Graseby Medical Ltd., Watford,

UK). Each time the button was pushed, the patient received

0.3 ml of the mixture, equivalent to 4.44 mg propofol and

0.039 mg alfentanil. No lockout period was used, which

resulted in a quantity of 22.2 mg propofol and 0.20 mg

alfentanil possible to give in 1 min.

For comparison and calculations the analgesic doses that of

alfentanil was multiplied by a factor of 7, resulting in

equipotent doses of morphine. Doses of fentanyl were multi-

plied by the factor of 100 [10].

2.3. Data analysis and statistics

Data are presented as either mean (S.D.) or median (inter-

quartile range). For the statistical analysis we used Statis-

tica1 Version 6.1 (Stat Soft, Inc., Tulsa, USA). To assess

differences between groups we used the Wilcoxon Matched

Pairs Test. Changes over time in surveillance data between

ACS and all PCS procedures were evaluated by repeated

measures ANOVA. Values of P < 0.05 were accepted as

significant.

Table 2 – Cardiopulmonary data and sedation data during AC

ACS

Mean (S.D.) Min–Max Me

SpO2, % 98 (3) 84–100 99

RR 14 (2) 8–17 16

PtcCO2 (kPa) 5.3 (0.4) 4.5–6.2 5

BIS index 84 (11) 48–99 93

HR 80 (11) 56–118 80

MAP (mmHg) 70 (15) 33–110 83

ACS = anaesthetist controlled sedation; PCS = patient controlled sedation

Values as mean and standard deviation (S.D.).

Lowest and highest values recorded are presented as Min and Max.

RR = respiratory rate, PtcCO2 = transcutaneous carbon dioxide, BIS index

3. Results

3.1. Details of patients (Table 1)

All the patients started the study with ACS followed by at least

two PCS. Ten of the 11 patients preferred to continue dressing

changes using PCS after the first ACS and PCS. One patient was

indifferent to the techniques but finally chose PCS.

3.2. Cardiopulmonary and surveillance data(Fig. 1 and Table 2)

Differences were found for SpO2, BIS, and transcutaneous

PCO2, between ACS and PCS. PCS gave higher mean SpO2

concentrations, but lower PCO2, and less sedation according to

BIS monitoring.

During ACS there was a slight decrease in saturation (<90%)

in two cases, the lowest respiratory rate was 8, and a lowest

BIS index was 48. Blood pressure (MAP) was reduced during

both techniques, the lowest values being between 50 and 68%

of baseline. Equivalent data during PCS showed no saturation

value lower than 94% and no respiratory rate less than 10. The

S and PCS

PCS Statistical differences (P)

an (S.D.) Min–Max

(2) 94–100 0.032

(5) 10–30 0.707

.2 (0.5) 4.0–6.2 0.004

(5) 67–99 0.027

(11) 65–102 0.686

(11) 58–115 0.663

.

= bispectral index, HR = heart rate and MAP = mean arterial pressure.

Fig. 3 – Capacity of the pump in comparison with the

patients with the highest demands. The doses are

presented accumulated from start to the last received

dose. Time (min) is from start.

Fig. 1 – BIS index during ACS (sedation controlled by

anaesthetist) and patient controlled sedation (PCS).

b u r n s 3 4 ( 2 0 0 8 ) 9 2 9 – 9 3 4932

lowest MAP recordings were within 60 and 96% of baseline.

Reduction in the BIS index also occurred during PCS with six

recordings less than 80.

3.3. Amount of drug given (Table 1 and Fig. 2)

During PCS all patient requested lower doses of propofol and

morphine (193.5 (83.5) and 12.1 (5.3) mg mean (S.D.)) than they

were given by the anaesthetist during ACS (394.8 (130.7) and

18.6 (9.0) mg). There was no difference in duration of

procedure between PCS (67.5 (19.0) min) and ACS (69.3

(26.7) min).

During PCS procedures, four patients had a third or less of

the required doses. The remaining seven patients had more

than 40% of required doses. Two of the four patients given less

than a third of the doses given were not satisfied with the

capacity of the pump, although PCS overall was preferred. The

nine remaining patients were satisfied with how they were

Fig. 2 – Doses of propofol given by anaesthetist (ACS)

compared with the doses requested by the patients (PCS)

during the first PCS.

given the doses. The capacity of the pump in relation to the

patients with high dose demands is shown in Fig. 3.

3.4. Procedure rating by personnel

After dressing changes, wound care personnel said that in all

cases, during ACS and PCS, sedation was adequate and, wound

care conditions was good. Two surgeons indicated, however,

that cleansing of the wounds could have been better on two

occasions during PCS.

3.5. Pain ratings (Table 3)

The highest mean (S.D.) pain ratings recorded, during wound

treatment were greater for PCS (4.9 (2.4)) than for ACS (1.5

(1,0)). Immediately and 10 min after dressing changes there

were no differences.

4. Discussion

The new finding of this study is that PCS, using a standard

technique with a fixed protocol comprising the drugs propofol

and alfentanil, can be used successfully as an alternative to

ACS for wound care in burned patients. When patients chose

between PCS and ACS, they chose PCS and truly preferred

being in charge of their sedation instead of relying on

somebody else. Although comparable data are lacking for

burns, previous studies indicated a preference for PCS during

other painful or unpleasant procedures [6,7]. A sense of

control, together with a more rapid and less unpleasant

recovery, together with the remaining possibility of deeper

sedation if necessary were reasons given for the choice of PCS

in the present study.

4.1. The procedure and pain

With a patient who can communicate, those who are caring

for the wounds informed them if unpleasant or painful

moments were to be expected, which gives them time to give

the sedation and analgesia properly as described previously

Table 3 – Intensity of pain evaluated with visual analogue scale (VAS)

Patient ACS PCS

VAS max VAS min 0 VAS min 10 VAS max VAS min 0 VAS min 10

1 2 4 4 5 3 0

2 0 7 3 8 5 3

3 0 4 1 4 1 1

4 0 0 0 1 0 0

5 2 0 0 3 0 0

6 2 3 1 6 1 1

7 2 0 0 3 0 0

8 2 6 3 8 4 2

9 3 2 2 5 4 2

10 2 2 2 3 3 1

11 1 1 0 8 2 0

Mean 1.45* 2.64** 1.45*** 4.91* 2.09** 0.91***

S.D. 1.04 2.42 1.44 2.39 1.81 1.04

ACS = anaesthetist controlled sedation; PCS = patient controlled sedation.

Values as mean and standard deviation (S.D.).

Maximum pain score (VAS max in table) represent highest pain experienced during changing dressings and VAS min 0 is the pain intensity

immediate after treatment. 10 min after dressing change pain intensity was again asked for (VAS min 10).

*P = 0.003; **P = 0.234; ***P = 0.109.

b u r n s 3 4 ( 2 0 0 8 ) 9 2 9 – 9 3 4 933

[6]. As this was the first series of PCS in this unit, the procedure

might be refined further as personnel get more acquainted

with the technique. We noticed that although special

emphasis was placed on preparing the patients and predicting

painful events, this was at times missed and may together

with the sometimes impatient behaviour of the personnel

explain the intermittently high VAS scores recorded during

care of the wound and PCS. Astonishingly, despite the high

VAS reported, it did not seem to affect the patients overall

judgement of the technique. To optimise the procedure,

therefore, further training may lead to less pain without the

need to change the doses of any drugs. The finding that the

length of the procedures done by ACS and PCS did not differ

also suggests that PCS was adequate.

We made an interesting, new discovery during this trial. In

all patients verbal complaints of pain (with corresponding

higher VAS scores) were not always accompanied by pressure

on the button to release the drug. There seemed to be

cognitive dysfunction with the present drugs and protocol

where, despite patients verbally complaining of pain, they

were unable to process that into the action of providing

themselves with more of the drug. No extra doses were given

by any bystander.

The high scores for pain recorded, and which seem

unreasonably high, argues for more analgesic. This could

theoretically be accomplished either by increasing the dose of

alfentanil given by PCS or by the supplementation of a longer-

acting analgesic at the start of the procedure. This must to be

examined further. We must point out that not a single patient

reached the maximum capacity of the pump; however, at the

same time it is difficult to give the maximum dose as there is a

hidden lockout time during which the pump delivers a dose

already requested. An alternative is to increase the volume of

alfentanil, but that will only be achieved with less propofol,

which will result in reduced sedation (given that the Graseby

pump as presently sold commercially delivers only a fixed

maximum volume). As far as we know patients appreciate the

present sedation. PCS itself is a factor in the satisfaction

among patients having cataract surgery, whether they did use

the pump with propofol or not [11] which in some way may

contribute to acceptance of the pain recorded in our study.

4.2. Safety

Despite the fact that this study used propofol and alfentanil in

concentrations higher than evaluated previously [6,7,11,12]

and a device without a lock out time, PCS was accomplished

without extreme values in cardiopulmonary or BIS data. The

patients were monitored closely and were given supplemen-

tary oxygen and it is therefore unlikely that any adverse

reaction has been overlooked. In all, we recorded no single

value that indicated compromise of breathing or circulation.

The fact that the corresponding decrease in the BIS index was

also minor supports this further. One must, however, be

cautious about the strength of this conclusion, as the number

of patients was small.

The smaller amounts of analgesic and sedative used in the

study is probably the explanation for there being fewer

symptoms of sedation after the procedure, which also

explains the lack of effects on circulation and breathing.

4.3. Sedation effects

BIS monitoring was used to acquire objective data about the

state of sedation, instead of using sedation scales, which often

involve some interactivity with the patient [7]. BIS indexes

were collected during the procedure with no major distur-

bances. The position of the patient’s head must be still and

controlled as the BIS monitor detects muscular movement,

and the signal quality may then be affected. We had few such

problems. BIS monitoring has been evaluated with propofol

sedation and is a good predictor of the level of sedation [13].

However, there is a lag time between decreasing values and

the outcome of an observer’s assessment [14]. As the purpose

b u r n s 3 4 ( 2 0 0 8 ) 9 2 9 – 9 3 4934

was not to delineate minor changes but rather to establish

sedation levels, mainly for the group using PCS, we do not

think that this has an impact on the interpretation of our

results. As for PCS, few values were recorded below 80%,

which is similar to the findings made by Coimbra et al. [9]. The

lowest values were present during ACS, which has been a

common finding by other authors who compared ASC and PCS

such as in, colonoscopy [6]. In that study the corresponding

figures were 58 (ACS) and 71 (ACS), respectively.

4.4. Other issues

There are some other issues of importance that should be

discussed.

Firstly, when we compare PCS and ACS we must

emphasise that the present standard in Sweden is to use

ACS for patients with major burns. We know that inter-

nationally many units do their dressings with no anaesthetist

support. We think that this argues for PCS as an interesting

alternative to the present procedure, in which anaesthetic

personnel are required, and it also may explain why patients

accept the procedure to such a large extent with not much

sedation. The PCS technique would certainly reduce con-

sumption of resource.

For PCS in the present study we used a push-button device,

which made it difficult to recruit patients with burnt hands.

These are common among our patients [15]. Several candi-

dates for the investigation were also excluded for this reason.

There are, however, other techniques that could be tried for

these patients.

The study was not blind; unfortunately this is not possible

with these techniques. We claimed that we used ‘‘true PCS’’, as

no ‘‘lockout’’ time was used. This is not completely true as in

practice there is a lockout time when the pump is giving the

actual dose. This time is in the range of 15 s, and may explain

the discrepancy between the doses requested and the doses

delivered.

In conclusion, PCS as presented in this study using a fixed

protocol appears an interesting alternative to ACS for dressing

changes in serious burns. The patients chose PCS, and

preferred it because they had better control and less sedation

after the procedure despite having less sedation during it and

more pain.

Conflict of interest

None of the authors have any financial or personal relation-

ship with people or organizations that could influence this

work. This study is performed on clinical basis of the Burn

Unit, with no external influence.

r e f e r e n c e s

[1] Latarjet J, Choiniere M. Pain in burn patients. Burns1995;21:344–8.

[2] Ashburn M. Burn pain: the management of procedure-related pain. J Burn Care Rehabil 1995;16:365–71.

[3] MacLennan N, Heimbach D, Cullen B. Anesthesia for majorthermal injury. Anesthesiology 1998;89:749–70.

[4] Jonsson CE, Holmsten A, Dahlstrom L, Jonsson K.Background pain in burn patients: routine measurementand recording of pain intensity in a burn unit. Burns1998;24:448–54.

[5] Alhashemi JA, Kaki AM. Anesthesiologist-controlled versuspatient-controlled propofol sedation for shockwavelithotripsy. Can J Anaesth 2006;53:449–55.

[6] Roseveare C, Seavell C, Patel P, Crisswell J, Kimble J, Jones C,et al. Patient-controlled sedation and analgesia, usingpropofol and alfentanil, during colonoscopy: a prospectiverandomized controlled trial. Endoscopy 1998;30:768–73.

[7] Oei-Lim VLB, Kalkman CJ, Makkes PC, Ooms WG. Patient-controlled versus anesthesiologist-controlled conscioussedation with propofol for dental treatment in anxiouspatients. Anesth Analg 1998;86:967–72.

[8] Knape JTA, Adriaensen H, van Aken H, Blunnie WP,Carlsson C, Dupont M, et al. Guidelines for sedation and/oranalgesia by non-anaesthesiology doctors. Eur JAnaesthesiol 2007;24:563–7.

[9] Coimbra C, Choiniere M, Hemmerling TM. Patient-controlled sedation using propofol for dressing changes inburn patients: a dose-finding study. Anesth Analg2003;97:839–42.

[10] Wood M, Alastair JJ. Drugs and anesthesia: pharmacologyfor anesthesiologist, 2nd ed., Williams & Williams; 1990.

[11] Herrick IA, Gelb AW, Nichols B, Kirkby T. Patient-controlledpropofol sedation for elderly patients: safety and patientattitude toward control. Can J Anaesth 1996;43:1014–8.

[12] Sim KM, Hwang NC, Chan YW, Sean CS. Use of patient-controlled analgesia with alfentanil for burns dressingprocedures: a preliminary report of five patients. Burns1996;22:238–41.

[13] Schraag S, Bothner U, Gajraj R, Kenny GN, Georgieff M. Theperformance of electroencephalogram bispectral index andauditory evoked potential index to predict loss ofconsciousness during propofol infusion. Anesth Analg1999;89:1311–5.

[14] Chen SC, Rex DK. An initial investigation of bispectralmonitoring as an adjunct to nurse-administered propofolsedation for colonoscopy. Am J Gastroenterol 2004;99:1081–6.

[15] Smith MA, Munster AM, Spence RJ. Burns of the hand andupper limb—a review. Burns 1998;24:493–505.