J Cancer Res Clin Oncol (2008) 134:489–493

DOI 10.1007/s00432-007-0311-0

ORIGINAL PAPER

The value of BAL Xuid LDH level in diVerentiating benign from malignant solitary pulmonary nodules

Ali Emad · Vahid Emad

Received: 13 December 2006 / Accepted: 27 August 2007 / Published online: 20 September 2007© Springer-Verlag 2007

AbstractPurpose Serum lactate dehydrogenase (LDH) concentra-tion is an indicator of tissue injury. It may be locallysecreted in some conditions. This study was performed inorder to investigate the value of LDH levels in bronchoal-veolar lavage Xuid (BALF) in the diVerentiation of a ben-ing, from a malignant solitary pulmonary nodules (SPN)and to assess its relationship with serum LDH levels.Methods The study was a prospective clinical study. Itincluded 59 patients with a SPN. They underwent bron-choscopy with bronchoalveolar lavage (BAL). Both totalserum and BAL LDH levels were measured.Results BALF LDH level was increased in all patientswith malignant SPN. The mean BALF LDH level was sig-niWcantly higher in patients with malignant SPN (342.23 §89.98) as compared to the benign ones (17.62 § 7.90)(P < 0.001). There was no correlation between BALF LDHand serum LDH level in patients with SPNs (P = 0.595).Conclusion BALF LDH levels are increased in patientswith malignant SPN, but had no signiWcant rise in benignSPN. This factor is useful in diVerentiating the benignSPNs from malignant SPNs.

Keywords LDH · Nodule · Malignant · Benign · BAL

Introduction

The Xexible bronchoscope has been in regular use for manyyears for investigating patients with solitary pulmonary nod-ules (SPN) (Torrington and Kern 1993). The most important

Wrst plan is to determine the likelihood of the nodule beingmalignant and then to decide whether the lesion should beremoved. The diagnostic yield of the standard combinationof wash, brush, and transbronchial biopsy (TBB) in patientswith SPN has a wide variability and depends on the location,size, character of the border, and the ability to do all sam-pling methods (Gasparini et al. 1995; Baaklini et al. 2000). Itoften requires the use of higher risk procedures, such as per-cutaneous needle biopsy and thoracotomy (Lillington 1991).

On the other hand, serum lactate dehydrogenase (LDH)concentration is an indicator of tissue injury, which may beincreased, in a variety of interstitial diseases (McFaddenand Oliphant 1991). Pulmonary alveolar proteinosis andPneumocystis carinii pneumonia are two well-documentedcauses of elevated serum LDH levels in the setting ofdiVuse lung disease (Zaman and White 1988; HoVman andRogers 1991). The BALF LDH level may be increased in avariety of interstitial lung diseases (Larivee et al. 1990).The increased level of this enzyme may reXect cytotoxicity(Hendersone et al. 1978). It is postulated that malignantlung lesion will cause cell damage. Therefore, elevatedBALF LDH levels are expected to be seen secondary to tis-sue damage (Drebt et al. 1996; Henderson et al. 1985).

To our knowledge, BALF LDH concentration has notbeen measured in cases with SPN. In the present series, wehave tried to investigate the value of BALF LDH levels fordiVerentiation between benign and malignant SPNs.

Methods

Patients

In our study, a SPN was deWned as a single, noncalciWed,well-circumscribed, spherical radiographic opacity that

A. Emad (&) · V. EmadPO Box 71345-1674, Shiraz, Islamic Republic of Irane-mail: lungdep@pearl.sums.ac.ir

123

490 J Cancer Res Clin Oncol (2008) 134:489–493

measures 1–4 cm in diameter and is surrounded completelyby aerated lung (Gambhir et al. 1998; Keith et al. 2002).There is no associated pulmonary inWltration, hilar enlarge-ment, atelectasis, or pleural eVusion. The patients with SPN,which is detected on screening chest radiograph, wereenrolled. The minimum size of the nodule was deWned as1 cm in diameter. The maximum size was deWned as 4 cm.Patients with known malignancy were excluded. In addi-tion, cases were excluded if they had hemoptysis, provencardiovascular diseases and abnormalities of liver functiontests. Patients with a history of cigarette smoking were alsoomitted because smoking may cause a possible inXamma-tory process and lung damage (Morrison et al. 1999; Li et al.1996). It has been proved that the gas phase of cigarettesmoke has many free radicals and oxidant molecules, withthe potential to produce additional oxidant molecules viaredox cycling, leading to an increased oxidative burden inthe lungs of smokers (Morrison et al. 1999). Therefore, cig-arette smoke inhalation leads to a depletion of antioxidants,release of inXammatory mediators, and an increase in epi-thelial permeability (Li et al. 1996). Only patients who had anormal endobronchial anatomy were included in the study.

The ethics committee of our Hospital Universityapproved the study protocol. Signed written informed con-sent was obtained from all cases prior to their enrolment.All patients underwent chest radiography, thoracic CTscanning, and abdominal sonography prior to any proce-dure. Complete cell count, blood chemistries, sputum cyto-logical, and sputum microbiology examinations wereobtained prior to bronchoscopy. Serum was drawn for totallactate dehydrogenase (LDH) level and albumin at the timeof bronchoscopy, prior to lavage. Bronchoscopy and bron-choalveolar lavage was performed in all eligible subjectand patients.

Bronchoalveolar lavage, transbronchial needle aspiration and transbronchial biopsy

Bronchoalveolar lavage (BAL) was performed using a Xex-ible Wberoptic bronchoscope (Olympus BF1T, Tokyo,Japan). It was done before any bronchial brushings. Afterlocal anesthesia, the bronchoscope was wedged for lavageinto a subsegmental involved bronchus determined by tho-racic CT scanning. Four 50-ml aliquots of sterile physio-logic saline solution, warmed at 37°C, were infused. Fluidwas immediately recovered by gentle suction after each ali-quot was introduced. Samples of the lavage Xuid weretaken for fungal, mycobacterial, and other bacterial cul-tures. In addition, the lavage Xuid was passed through twosheets of gauze to eliminate mucus and stored at 0°. Onesmall aliquot of this Xuid was utilized to count the total cellnumber, and another was spun in a cytometer at 400 rpmfor 10 min. The cell pellet was washed once in Hanks' bal-

anced salt solution (without calcium and magnesium).May-Grünwall-Giemsa stains smear served to identifydiVerential proWles after cytospin preparation by counting600 cells. Total cell counts were determined with a hemo-cytometer. Macrophages, lymphocytes, and neutrophilswere counted, and results were expressed as percentages.Unconcentrated supernatant was frozen at ¡20°C beforeLDH and albumin were measured.

Total LDH concentration was determined on the super-nant. The rate of reduction of nicotinamide adenine dinucleo-tide (NAD) to NADH in the presence of L-lactate wasutilized to quantitate total LDH activity (Bergmeyer and Ber-net 1974). It was reXected by the rate of absorbance increaseat 340 nm. All BAL LDH results were directly expressed inmilliinternational units (mIU) and were also given as mIU/mg albumin when albumin was used for standardization.Albumin concentration in serum was measured by immuno-nephelometry (Delacroix et al. 1985). The technique ofimmunoradiometric assay was used to quantitate the albuminconcentration in the BAL Xuids (Delacroix et al. 1982).

The peripheral nodule was approached Wrst by brush andthen by transbronchial needle aspiration (TBNA) and trans-bronchial biopsy (TBB) under Xuoroscopic guidance. Ofcourse, in some cases the approach was diYcult.

Thoracotomy was used when procedures such as sputumcytology, sputum microbiology, and bronchoscopy (bron-chial washing, brush, TBNA, TBB) fail to provide adequateinformation.

Control cases

All subjects gave informed consent. Bronchoscopy withBAL and subsequent measurement of LDH and albuminwere performed for 21 non-smoker healthy adult volunteerswho acted as our control group, using the same methodsdescribed above. Serum LDH and albumin concentrationswere measured immediately prior to BAL.

Statistics

All results are mean § SD unless otherwise indicated. TheStudent's t test, Chi-square and Kruskal–Wallis were usedfor analysis of data. The Spearman rank test was performedto test for a correlation between the patients, BAL LDH lev-els and the serum LDH concentrations. The sample sizewas enough according to our statistics.

Results

Thirty-Wve males and 24 females with an average age of46.88 § 9.19 years (range 18–63 years) were examined.The age range of the control groups was 29–43 years

123

J Cancer Res Clin Oncol (2008) 134:489–493 491

(mean § SD = 33.57 § 4.22). There was a signiWcantdiVerence in the mean age of the patients with malignantand benign nodules (P = 0.03). With respect to age, the twostudy patients signiWcantly diVer from the control group(P < 0.005). In 49 patients (83%) there were no symptomsascribable to the SPN; in the remaining, 10 cases had cough(n = 7), and chest pain (n = 3).

The localization of the abnormality was the right middlelobe in three patients (5.1%), the left upper lobe in eleven(18.6%), the right upper lobe in 13 (22.1%), the left lowerlobe in 17 (28.8%), and the right lower lobe in 15 (25.4%).Table 1 shows the Wnal etiological diagnoses. In 17 patients(28.8%), the histology was benign, and in 42 (71.2%), thespecimens were malignant. Diagnosis was obtained by pre-bronchoscopic sputum specimens in three patients (5%), bybronchoscopic procedures in 36 patients (61%) and by tho-racotomy in 20 (34%).

The total volume of Xuid retrieved and the total numberof cells recovered were not signiWcantly diVerent betweenthe patients with SPN and the control subjects (P = 0.61)(Table 2). The percentages (mean § SD) of macrophages,lymphocytes, and granulocytes were signiWcantly higher inpatients than the control group (P = 0.0008, 0.03, and 0.01,respectively, Table 2).

For proportions of macrophages and lymphocytes, therewere no signiWcant diVerences between patients with abenign nodule and those with a malignant lesion (P = 0.34and P = 0.72) (Table 3). The proportion of granulocytes

was signiWcantly higher in patients with a benign lesionthan in cases with malignancy (P = 0.01).

The mean BAL Xuid and serum LDH levels(mean § SD) of all patients and control subjects are shownin Table 4. The highest BAL LDH level was 26 mIU/ml(range 4.60–26) in control group, 83 mIU/ml (range 6–83)in patients with benign nodule and 147 mIU/ml (range 33–147) in those with malignant nodule. Overall, the meanLDH level of BAL in patients with malignant pulmonarynodule was statistically higher than in the other two groups(the patients with benign nodule and the healthy controls)(P < 0.0001). No signiWcant diVerence between the abso-lute value of BAL LDH level in the patients with benignpulmonary nodule and the control subjects was found(P = 0.12). In patients with malignant pulmonary nodule,no signiWcant correlations between BAL Xuid LDH levelsand the proportions of macrophages (� = 0.03, P = 0.80),lymphocytes (� = 0.11, P = 0.44, and granulocytes weredetected (� = 0.21, P = 0.16). No statistical signiWcant cor-relation between BAL Xuid LDH levels and the size of themalignant nodules was found (P = 0.63) too.

In two patients with a benign nodule and one patientwith a malignant nodule, the serum level of LDH wasincreased (>250 mIU/ml). None of the control group casesshowed an elevated serum LDH level. There was no signiW-cant correlation between BAL LDH levels and serum LDH

Table 1 Final diagnosis in all patients with SPN

Patients with SPN (n = 59)

Squamous cell carcinoma 12

Adenocarcinoma 21

Large cell carcinoma 2

Small cell carcinoma 6

Metastatic disease 1

Hamartoma 4

Tuberculosis 2

Sarcoidosis 2

NonspeciWc 9

Table 2 Bronchoalveolar lavage cell content in patients with SPNsand normal control subjects

Patients with SPN (n = 59)

Control subjects (n = 21)

P value

Fluid recovery, ml 163.18 § 14.57 165.42 § 11.57 0.61

Total cells, £ 106 86.44 § 8.44 85.90 § 7.56 0.74

Macrophages (%) 89.84 § 2.45 92.10 § 2.07 0.0008

Lymphocytes (%) 6.40 § 2.65 5.15 § 2.04 0.03

Neutrophils (%) 2.96 § 1.17 2.29 § 1.18 0.01

Table 3 Bronchoalveolar lavage cell contents in patients with a be-nign and a malignant SPN

Patients with a benign nodule(n = 17)

Patients with a malignant nodule(n = 42)

P value

Fluid recovery, ml 163.88 § 14.83 162.90 § 13.42 0.56

Total cells, £ 106 84.41 § 9.48 87.26 § 7.97 0.30

Macrophages (%) 90.474 § 2.30 89.59 § 2.49 0.34

Lymphocytes (%) 6.65 § 2.49 6.30 § 2.18 0.72

Neutrophils (%) 2.32 § 1.02 3.22 § 1.30 0.01

Table 4 Serum and BAL LDH levels (mIU/ml) in cases of controlgroup and all patients with benign and malignant pulmonary nodules

*P < 0.0001 (between malignants and benigns), **P < 0.0001(between malignants and controls)

BAL LDH (mIU/ml)

Serum LDH (mIU/ml)

Control (n = 21) 12.16 § 6.18 146.23 § 25.93

All Patients with SPN (n = 59) 66.66 § 39.89 135.78 § 57.32

Patients with a benign nodule (n = 17)

19.08 § 18.35 147.11 § 64.83

Patients with a malignant nodule*, ** (n = 42)

85.92 § 28.31 131.19 § 54.15

123

492 J Cancer Res Clin Oncol (2008) 134:489–493

levels (� = 0.01, P = 0.94) in patients with a malignant nod-ule. The ratio of serum to BAL LDH level ranged from 0.53to 3.65 (mean § SD 1.71 § 0.87) in patients with malig-nant nodules, from 1.26 § 27.50 (mean § SD 11.25 §7.12) in patients with benign nodules and from5.59 § 33.91 (mean § SD 14.93 § 7.24) in control group.The diVerence between the two group patients was quitesigniWcant (P < 0.0001), but no signiWcant diVerencebetween the patients with benign nodules and the controlswas found (P = 0.12).

Table 5 shows albumin levels and LDH/albumin ratiosof BAL and serum in all patients and control subjects.

Discussion

Although most SPNs are benign, as many as one third rep-resent primary malignancies, and nearly one quarter may besolitary metastases (Leef and Klein 2002; Yankelvitz andHensche 2000; Greenlee et al. 2001). The prevalence ofmalignancy in nodules varies widely, depending on thepatient populations. Older age, a history of cigarette smok-ing, and a past history of cancer all increase the probabilitythat an SPN is malignant (Swenson et al. 1997). Benignsolitary pulmonary nodules are more common in the youngand in nonsmokers (Libby et al. 1995). Therefore, SPNsshould be diagnosed, treated, and managed in order to missany treatable cancer.

Current usual methods to diagnose SPNs include sputumcytology, Wne needle aspiration (FNA), bronchoscopy withbrushings, washings and bronchial biopsy, and Wnally themost invasive technique of open lung biopsy (Baakliniet al. 2000; Johnston and Elson 1991). Bronchoscopy maymostly help in patients with central nodules and less inperipheral ones.

On the other hand, the BAL Xuid LDH level mayincrease in a variety of interstitial lung diseases. Theincreased level of this enzyme may reXect cytotoxicity(Robbins et al. 1992; Danpure 1984). According to thisfact, the fundamental question in the present study iswhether the determination of the BAL Xuid LDH level canguide the physician to improve their diagnosis, especiallyin complex cases.

To the best of our knowledge, this is the Wrst reportregarding the measurement of LDH in the BALF of patientswith SPN. All benign SPN cases except one case of tuber-culosis (34 mIU/ml) and another case of sarcoidosis(83 mIU/ml) showed a BALF LDH level of less than32 mIU/ml. On the other hand, among the malignant SPNpatients, all except one with adenocarcinoma showed aBALF LDH level of more than 33 mIU/ml as compared tobenign SPN cases. The mean BALF LDH level was signiW-cantly higher in patients with malignant SPN than that ofbenign ones too.

The elevation of serum LDH levels in some benign SPNcases may be due to other associated causes rather thansimply due to lung injury. Other causes for an increase inserum LDH level, such as any hepatic disease and otherorgan abnormalities, were excluded in our patients. One ofall malignant SPN patients had an increased serum LDHlevel. The Wnding of serum to BAL LDH ratio was interest-ing. All malignant SPN cases had a value less than 3,except one benign SPN case who had a sarcoid nodule. Theincreased level of the lavage Xuid albumin is also sugges-tive of an increased permeability of the alveolocapillarymembrane.

The study shows that the ratio of LDH/albumin BALwas signiWcantly higher than the ratio of LDH/albuminserum in all benign SPN cases and controls. Meanwhile, theserum/lavage LDH ratio in control subjects was signiW-cantly higher than in malignant SPN cases (P < 0.0001).

Consideration this fact that the rate of rise of the BALXuid LDH (molecular weight = 140,000 Da), however,seemed to be higher than that of its albumin concentration(molecular weight = 69,000 Da), therefore, it may be postu-lated that the elevated BAL LDH level is mainly causedfrom its local production secondary to tissue damage, ratherthan reXecting passive leakage from the circulation to alve-oli.

No correlation was observed between the BAL andserum LDH levels in malignant SPN cases. Meanwhile as itwas indicated, the serum LDH level was not increased in allmalignant SPN cases, except in one case whose nodulediagnosed as metastatic disease. Therefore, this study maysuggest that the elevated total serum LDH in this case maybe a reXection of lung tissue injury. In other words,

Table 5 Albumin levels and LDH/albumin ratios of BAL and serum (mean § SD) in control group and all patients with benign and malignantpulmonary nodules

Serum BAL

Alb (�g/ml) LDH/Alb (mlU/mg) Alb (mg/ml) LDH/Alb (mlU/mg)

Control 37.28 § 3.84 3.95 § 0.79 22.01 § 18.30 572.17 § 326.06

Cases with a benign nodule 37.82 § 3.87 3.88 § 1.56 28.38 § 12.85 755.65 § 741.60

Cases with a malignant nodule 37.76 § 3.63 3.46 § 1.34 35.44 § 17.75 2,889.27 § 1,411

123

J Cancer Res Clin Oncol (2008) 134:489–493 493

increased backXow of the LDH molecule produced withinthe alveoli into the circulation through the alveolocapillarybarrier in the presence of an inXammatory state may lead tothe elevated serum LDH level in this case with metastaticdisease. Of course, it should be investigated more.

Although the problem of measuring non-cellular compo-nents in BAL has never been solved and the normal rangefor BAL LDH concentration has not been clearly estab-lished, yet, a very low value (possible less than 30 mIU/ml)may help to exclude the diagnosis of malignant nodule. Inaddition, a serum to BAL LDH ratio of more than 3.5 mayhighly reject the diagnosis of malignant nodule too. Furtherinvestigations are needed for a more clear elucidation ofthis issue.

In conclusion, determination of the BAL Xuid level inXexible bronchoscopy may be useful in the diagnosis ofperipheral pulmonary nodules eluding diagnosis by othertechniques. A low BAL Xuid LDH level in a patient withSPN who does not have a tissue diagnosis may be deemedacceptable for observation and follow up. This may savepatients the need for operative procedures. We encouragefurther studies addressing the improvement of this methodin the evaluation of SPNs.

We try to extend our survey in order to Wnd the realvalue of measuring BAL LDH in the management ofpatients with single pulmonary nodule.

References

Baaklini WA, Reinoso MA, Gorin AB, Sharafkaneh A, Manian P(2000) Diagnostic yield of Wberoptic bronchoscopy in evaluatingsolitary pulmonary nodules. Chest 117:1049–1054

Bergmeyer HU, Bernet E (1974) Lactate dehydrogenase UV-assaywith pyruvate and NADH In: Bergmeyer HU (ed) Methods ofenzymatic analysis. vol 2. Academic, New York, pp 574–579

Danpure CJ (1984) Lactate dehydrogenase and cell injury. Cell Bio-chem Funct 2:144–148

Delacroix DL, Hodgson HJF, McPherson A, Dive C, Vaerman JP(1982) Selective transport of polymeric immunoglobulin A inbile: quantitative relationships of monomeric and polymericimmunoglobulin A, immunoglobulin M and other proteins inserum, bile and saliva. J Clin Invest 70:230–241

Delacroix DL, Marchandise F, Francis C, Sibille Y (1985) Alpha-2-macroglobulin, monomeric and polymeric immunoglobulins A,and immunoglobulin M in bronchoalveolar lavage. Am RevRespir Dis 132:829–835

Drebt M, Cobben NA, Henderson RF et al (1996) Usefulness of lactatedehydrogenase and its isoenzymes as indicators of lung damageor inXammation. Eur Respir J 9:2736–2742

Gambhir SS, Shepherd JE, Shah BD et al (1998) Analytical decisionmodel for the cost-eVective management of solitary pulmonarynodules. J Clin Oncol 16:2113–2125

Gasparini S, Ferretti M, Bichi Secchi E, Baldelli S, Zuccatosa L,Gusella P (1995) Integration of transbronchial and percutaneousapproach in the diagnosis of peripheral pulmonary nodules ormasses. Experience with 1027 cases. Chest 108:131–137

Greenlee RT, Hill-Harmon MB, Murray T et al (2001) Cancer statis-tics, 2001. CA Cancer J Clin 51:15–36

Hendersone RF, Damon EG, Henderson TR (1978) Early damageindicators in the lung. I. Lactate dehydrogenase activity in theairways. Toxicol Appl Pharmacol 44:291–297

Henderson RF, Benson JM, Hahn FF et al (1985) New approaches forthe evaluation of pulmonary toxicity: bronchoalveolar lavageXuid analysis. Fundum Appl Toxicol 5:451–458

HoVman RM, Rogers RM (1991) Serum and lavage lactate dehydroge-nase isoenzymes in pulmonary alveolar proteinosis. Am RevRespir Dis 143:42–46

Johnston W, Elson C (1991) Respiratory tract. In: Bibbo M (ed) Com-prehensive cytopathology, WB Saunders, Philadelphia, pp 320–398

Keith CJ, Miles KA, GriYths MR et al (2002) Solitary pulmonary nod-ules: accuracy and cost-eVectiveness of sodium iodide FDG-PETusing Australian data. Eur J Nucl Med 29:1016–1023

Larivee P, Cantin A, Dufresne A, Begin R (1990) Enzyme activities oflung lavage in silicosis. Lung 168:151–158

Leef JL 3rd, Klein IS (2002) The solitary pulmonary nodule. RadiolClin North Am 40:123–143

Li XY, Rahman I, Donaldson K, MacNee W (1996) Mechanisms ofcigarette smoke induced increased airspace permeability. Thorax51:465–471

Libby DM, Henschke CI, Yankelevitz DF (1995) The solitary pulmunodule: update 1995. Am J Med 99:491–496

Lillington GA (1991) Management of solitary pulmonary nodules. DisMon 37:271–318

McFadden RG, Oliphant LD (1991) Serum lactate dehydrogenase ininterstitial lung disease (letter) Chest 100:1182

Morrison D, Rahman I, Lannan S, MacNee W (1999) Epithelial per-meability, inXammation, and oxidant stress in the air spaces ofsmokers. Am J Respir Crit Care Med 159:473–479

Robbins RA, Koyama S, Spurzem JR, Rickard KA, Nelson KJ, Goss-man GL et al (1992) Modulation of neutrophils and mononuclearcell adherence to bronchial epithelial cells. Am Rev Respir Dis7:19–29

Swenson SI, Silverstein MD, Ilstrup DM et al. (1997) The probabilityof malignancy in solitary pulmonary nodules: application to smallradiologically indeterminate nodules. Arch Intern Med 57:849–855

Torrington KG, Kern JD (1993) The utility of Wberbronchoscopy in theevaluation of solitary pulmonary nodules. Chest 104:1021–1024

Yankelvitz DF, Hensche CI (2000) Lung cancer: small solitary pulmo-nary nodules. Radiol Clin North Am 38:1–9

Zaman MK, White DA (1988) Serum lactate dehydrogenase levelsand Pneumocystis carinii pneumonia: diagnosis and prognosticsigniWcance. Am Rev Respir Dis 137:796–800

123