- Email: [email protected]
Pulmonary damage after modest exposure to zinc chloride smoke
RESPIRATORY
MEDICINE
(1999) 93,885-890
Pulmonary damage after modest exposure to zinc chloride smoke B. ZERAHN*,A. KOFOED-ENEVOLDSEN~, B. V. JENSEN*, J. MQLVIG?, N. EBBEHDJ~, J. S. JOHANSEN~ AND I.-L. KANSTRUP* *Department of Clinical Physiol. and Nuclear Medicine, Herlev Hospital, University of Copenhagen, ?Department of Endocrinology, Herlev Hospital, University of Copenhagen,SDepartment of Occupational and Environmental Medicine, Bispebjerg Hospital, University of Copenhagen and §Department of Medicine, Division of Rheumatology, University of Copenhagen, Hvidovre Hospital, Copenhagen, Denmark Thirteen soldiers (11 men and two women) were exposed to zinc chloride smoke (ZCS) during a combat exercise. Even though their initial symptoms were modest, a prolonged follow up with lung function testing and blood samples was undertaken due to previous caseswith fatal outcome after exposure to ZCS. Four weeks after exposure there were statistically significant declines from baseline values in lung diffusion capacity and total lung capacity of 16.2% and 4.3%, respectively. At the same time plasma levels of fibrinogen and zinc were significantly elevated, though mainly within the normal range. All variables showed a tendency towards normalization at follow up 8 weeks and 6 months after exposure. These findings indicate an unexpected quantifiable damage to lung parenchyma with a remarkable delay after modest exposure to zinc chloride smoke despite sparse initial symptoms. Exposure to high concentrations of ZCS may induce adult respiratory distress syndrome (ARDS) after a symptom free period of up to 12 days from exposure. Even though none of the soldiers in the present study developed ARDS the assessmentof lung diffusion capacity and acute phase reactants is proposed as a supplement when monitoring patients after exposure to ZCS. RESPIR. MED.
(1999) 93, 885-890
0 -_I_---_-._-
Introduction Zinc chloride smoke (ZCS) screens are used for both military and civilian screening purposes. The use of these smoke bombs in confined spaces has been widely abandoned due to several lethal incidents (l-3). However, the use of ZCS in open air is considered safe within certain limitations (4,5). Immediate symptoms after exposure may be irritation of the eyes and the upper respiratory tract with lachrymation, metallic taste, coughing, hoarseness and soreness of the throat, which in mild cases last from a few hours to a few weeks (6,7). In more severe cases nausea vomiting, chest pain, and headache also occur (8,9). Clinical and paraclinical findings include cyanosis, pyrexia, leucocytosis (mostly polymorfonuclear leucocytes), reduced vital capacity (VC) and forced expiratory volume in 1 set (FEV1) at the peak of symptoms with recovery during convalescence, increased zinc urinary excretion and Paperreceived7 May 1999and accepted18 May 1999. Correspondence should be addressed to: Bo Zerahn, MD, Department .of Clinical Physiology and Nuclear Medicine, FrederiksbergHospital, 2000 Frederiksberg,Denmark. Fax: +45 39 77 76 22; E- mail: [email protected] 0954-6111/99/120885+06 $12*00/O
-_l(--.-.--_
1999 HARCOURT
PUBLISHERS
LTD
- -- __.-“,“_x.-..., I .,~..-‘~.--~.-I__--i-..~.--^---
.-^I”~-
various degrees of hypoxemia. Chest radiographs may show diffuse infiltration that in some cases persist in various degrees without clinical symptoms. In more severe cases emphysematous bullae and pneumothorax occur (10). Exposure to high concentrations may induce adult respiratory distress syndrome (ARDS) after a symptom free period of up to 12 days from exposure (11,12). The outcome is frequently fatal, and in such cases autopsies have shown extensive interstitial and intra-alveolar space fibrosis and arterial and venous lumen reduction as well as obliteration and widespread occlusion of microvessels. However, the exact mechanism by which ARDS is triggered is unknown (13). In exceedingly high concentrations of ZCS death is caused within hours by oedema of the larynx and spasm of the glottis (14). In lesser but still very high concentrations severe haemorrhagic ulceration and excoriation of the upper respiratory tract may lead to a fatal outcome (1). At present, only casuistic reports have dealt with diffusion capacity after ZCS exposure (11,15,16). We describe a 6-month follow up of lung function variables and specific serum variables in 13 soldiers, who were exposed to modest concentrations of ZCS during a routine military exercise. 0
1999 HARCOURT
PUBLISHERS
LTD
886
B. ZERAHN
ETAL.
Material and methods PATIENTS Thirteen soldiers, (two women and 11 men) aged 19 to 26 (median age 20 years) were exposed to ZCS during a combat exercise. Eleven of the soldiers were never smokers while the remaining two were present smokers (4.25 and 6 pack-years, respectively). The soldiers were together with nine unexposed colleagues either defending or attacking a combat exercise building with a wide door and window openings. A total of two smoke canisters were ignited in the open air, but due to wind conditions the smoke drifted into the combat exercise building through the open windows and doors. Two of the soldiers inside the building were exposed for 5-10 min without airway protection. The remaining 11 soldiers were briefly exposed while running through the plume of smoke and/or upon entering the combat exercise building. Primarily 12 of the soldiers were hospitalized on the day of exposure for observation due to soreness of the throat, coughing, chest pain and mild headache. None of the soldiers had severe respiratory symptoms when admitted, but four of them complained of dyspnoea during the combat exercise. Upon admittance all soldiers were treated with inhalation of 1 mg budesonide and 200 mg intravenous hydrocortisone followed by 75 mg oral prednisolone after 12 h. Four soldiers were then treated with oral prednisolone 40 mg daily for 1 week and then gradual reduction in dose with Smg every third day. Three of the soldiers were treated because of the duration and concentration of smoke exposition and the fourth was treated because of persistent coughing and chest pain. This soldier had a laryngoscopy 5 days after exposure, showing signs of hypopharyngeal inflammation interpreted as chemical pharyngitis. One soldier was treated for 18 h with intravenous N-acetylcysteine (140 mg kg-’ day-‘) until blood samples were obtained showing normal values for plasma zinc. The thirteenth soldier was admitted to hospital the day after exposure due to mild initial symptoms. He was not treated with steroids but joined the follow-up programme.
METHODS All soldiers had spirometry and lung diffusion capacity testing performed using a Pulmonary Function Testing System 1070 (Medical Graphics Corp. Minnesota, U.S.A.). One soldier however, had her first lung function test performed at another hospital in Copenhagen county. The results from the latter test have been used as her baseline values and have not been presented in the preliminary study on our results (17). All soldiers had the first lung function test upon the first day after exposure. The following pulmonary variables were recorded at each session: Vital capacity (VC), forced expiratory volume in 1 set (FEVt), maximal expiratory flow when xx% of the FVC remains to be delivered (MEF,), total lung capacity (TLC), lung diffusion capacity with adjustment for actual
haemoglobin content (DLCO,,,) and diffusing capacity per alveolar volume (DL/Va) were recorded (18). TLC and DLCOcorr were determined with the single breath technique after inhalation of a gas mixture containing carbon monoxide and neon with the latter as an indicator of the dilution. The better of two results that differed no more than 3% was used as the final test result. Only experienced lung function technicians performed the tests. Lung diffusion capacity was adjusted for haemoglobin level according to Cotes et al. (19). Blood samples were obtained from all soldiers on the second day and subsequently after 1, 2,4, and 8 weeks. At day 0 blood samples were obtained from 10 of the soldiers and 29 weeks after exposure from 11 of the soldiers. From these blood samples plasma levels of zinc, fibrinogen and Creactive protein were determined along with haemoglobin, blood sedimentation rate and differentiated while blood cell counts. Also serum levels of YKL40 (human cartilage glycoprotein-39) and the aminoterminal propeptides of type 1 and type III collagen (PINP and PIIINP) were determined on each blood sample. YKL40 is an 18glycosyl hydrolase that has been suggested to function in the degradation of extracellular matrix or tissue inflammation (20,21). Both PINP and PIIINP are regarded as markers of formation of collagen type I and III respectively (22,23). However, blood samples were only sufficient for determination of YKL40, PIIINP and PINP at day 0 and day 1 for 10 of the soldiers and from day 2 for 12 of the soldiers. Serum levels of YKL-40 and PIIINP were determined by radio immunoassays (20,23) and PINP by enzyme linked immunosorbent assay (22). Radiographs of the thorax and arterial blood gas analyses were also obtained from day 1 after exposure and during follow-up.
DATA ANALYSIS Changes in lung function variables and blood analyses with time were tested using the Wilcoxon non-parametric test for paired data and the Friedmann non-parametric test for several related samples.
Results Data regarding changes in lung function variables for the 13 exposed soldiers are listed in Table 1. All baseline values for lung function were within normal range when compared to a Danish reference population (24). Significant declines in DLCO,,,, TLC, and Dl/Va.were found, with the lowest values observed 4 weeks after exposure. MEFzs was also lower 4 weeks after exposure compared to baseline values, but the Friedmann test only showed a trend towards significance. The development of changes in DLCO,,, compared to initial values is illustrated in Fig. 1. A significant increase in VC was found 2 weeks after exposure. During follow-up we observed no significant drifts or jumps with time during routine calibration tests of the pulmonary test system. We were not able to demonstrate any correlation between the degree of exposure,
1. Lung function 0 (25-75%)
(4.5-8.5) (3.5-6.4) (1.3-2.0) (2.9-5.3) (6.3-l 1.6) (8.5-17.3) (2.1-3.0)
7.0 5.4 1.6 4.6 8.6 14.2 2.4
after mild exposure
median
variables
7.1 5.4 1.7 4.7 8.2 13.2 2.3
1 (25-75%)
(4.48.8) (34-6.7) (0.8-2.3) (3.0-5.6) (6.1-10.6) (8.33155)* (1.8-2.7)***
median
to zinc chloride
7.3 5.7 2.0 4.6 8.0 12.8 2.2
2 (25-75%)
(3.8-9.2) (3.5-6.7)*** (0.8-2.3) (3G6.1) (5.3-10.6)(*) (8.2-15.1)*** (1.7-2.5)***
median
smoke
6.7 5.5 1.6 4.6 7.5 11.9 2.1
4 (25575%)
(4.3-9.0)** (3.66.5) (1.G2.0) (2.9-6.0) (5.7-9.5)** (7.3-13.7)*** (1.7-2.4)***
median 7.0 5.5 1.9 4.7 8.3 12.7 2.1
median 6.8 5.6 1.8 4.5 8.2 13.7 2.2
8 (25-75%)
(44-9.0) (3.66.5) (l&2.1) (2.8-6.0) (5.7-9.9) (7.3314.9)*** (l&2.5)***
median
(4.8-8.6)* (3&6.8)* (1.3-2.1) (3.1-6.1) (6.3-10.1) (7&14.3)* (1.8-2.8)
29 (25-75%)
21.4*** 20.2*** 10.4(*) 8.6 9.5(*) 28.6*** 26.6***
Friedman test Chi-square
Significance levels are indicated by asterixes: (*)O.l > P > 0.05;* P < 0.05;** P-c 0.02; *** P
TLC(l) vc (1) RV (1) FEVl (1) ME& (l/s) DLCO,,,, (mM/min/mmHg) DL/Va (mM/min/mmHg/l)
Weeks after exposure
TABLE
888
B.
ZERAHN
ETAL.
from baseline values were found during the remaining follow-up period. For all soldiers both plasma levels of PIIINP and PINP declined from day 0 to day 1 by median 32%, PcO.01, n= 10. No significant changes with time for either of the two collagen formation markers were found hereafter. Chest radiographs and arterial blood gas analyses remained normal for all the soldiers during follow-up. * **
**
16 -
Discussion
14 12 --
-
10 86-
18 -
81
-
-
0.3
o
: IIII1IIII
.‘,
012
4 Weeks
8 after mild exposure
I
29
to zincchloridesmoke 1. Changes from values at day of DLCO,,,, in percent and of plasma zinc and fibrinogen in ,u mol-’ after modest exposure to zinc chloride smoke. *P
FIG
**p
***p
treatment with steroids, or smoking and the changes in lung function variables. None of the soldiers had leucocytosis upon admittance. But the initial levels of both neutrophils and monocytes were significantly higher than 29 weeks after exposure (36% and 25% respectively, P-c 0.01). Only the four soldiers who received oral prednisolone had levels of neutrophils above normal range during follow-up. Plasma levels of fibrinogen were significantly elevated during a period from 1-8 weeks after exposure compared to initial values and some peak values exceeded normal range (Fig. 1). Plasma levels of zinc were above initial values at 28 weeks after exposure (Fig. 1). Individual peak levels for plasma zinc were dispersed over a wide range of time from the second day to 8 weeks after exposure and remained within normal range. Five out of the 13 soldiers had slightly elevated plasma levels of CRP and eight out of 13 had a slight increase in blood sedimentation rate during the period from 2-8 weeks after exposure. We were not able to demonstrate a correlation between changes in lung function variables and changes in plasma levels of zinc or fibrinogen. Plasma level of YKL-40 increased by median 1118% from day zero to day two (n = 10, P= 0.005). No differences
We have demonstrated a significant decreasein DLCO,,,,, TLC and Dl/Va after modest open-air exposure to ZCS with a corresponding increase in serum levels of fibrinogen and zinc. These findings indicate that sufficient amounts of ZCS must have passed deeply enough into the lower respiratory tract to induce a quantifiable reduction in lung function. Preliminary data and statistics on changes in DLCO,,,, after exposure to ZCS from the present study have already been published (17). The present study is an expanded version with prolonged follow-up, improved computed statistical analyses and extended blood sample analyses. Gas transfer for carbon monoxide after exposure to ZCS has only been described previously in three singular cases. One was found normal (1 l), another had a slight reduction in DLCO 4 months after exposure (15) and the third one showed moderate to severe reduction 2 weeks after exposition with recovery after 2-3 months (16). A ZCS bomb contains hexachlorethane (C&), zinc oxide (ZnO) and calcium silicide (CaS$ along with minor amounts of aluminium, magnesium and ammonium chloride. When ignited these components form mainly zinc chloride (ZnClz), which again in aqueous solution forms hydrochloric acid (HCl) and zinc oxychloride (ZnClOH) (4). These three chemicals are all corrosive. Also minor amounts of phosgene (COCQ carbon monoxide (CO) and polyaromatic hydrocarbons are produced, but the pulmonary and toxic impact of these compounds is generally considered to be of little or no importance in this context (25). In confined spaces an ignited smoke bomb rapidly absorbs all available moisture from the air. In such a case the main component of the smoke will be zinc chloride in minute sized particles ranging from 0.01 to 25~ (7) which may pass deep into the respiratory tract (26). With increasing air humidity successivelylarger amounts of zinc chloride will be hydrolysed into hydrochloric acid and zinc oxychloride. The hydrochloric acid absorbs comparably more water and forms larger particles mainly deposited in the upper respiratory tract (27). In rats ZCS induces pulmonary oedema, alveolitis and subsequently fibrosis (28). The pathological findings after fatal cases in man are similar. They include extensive interstitial and intra-alveolar fibrosis, diffuse microvascular obliteration and widespread occlusion of the pulmonary arteries (29). The progression of ARDS is suggested to be mediated through a programmed release of mediators from recruited and activated inflammatory cells and stimulation
ZINC
of mediator cascades that include cytokine interactions (13). The increase in serum levels of fibrinogen and zinc and the decline in ,DLCO,,,, TLC, Dl/Va and to a lesserextent MEFz5 is compatible with an ulcerative and fibrotic process in the lung alveoli and smaller bronchioli with a delayed onset and an apparent maximum at 4 weeks after exposure (30). However, the modest symptoms of the soldiers did not justify attempts to verify this by histology neither by biopsy nor by bronchoalveolar lavage. The following healing response seemsslow, and one cannot rule out the possibility of permanent damage from the present data. The slight increase in VC and FVC found 2 weeks after exposure can be explained by adaptation to the test procedure. It is also noteworthy that the reduction VC and FEVr described in patients with heavy exposure to ZCS was not demonstrable in the present material. This may indicate that DLCO,,,,, TLC and Dl/Va are more sensitive procedures in such cases with mild exposure. Although a sufficiently matched control group was not obtainable due to the acute circumstances of the study, we are confident that the observed changes are not due to circadian variations or lung function technicians. YKL40 is presumed to function in tissue remodelling and serum YKL40 levels are elevated in patients with active rheumatoid arthritis (20) in patients with liver fibrosis (21) and in patients with recurrent breast cancer and metastases to bone and viscera (31). YKL-40 is secreted by human macrophages (32) and is a matrix protein in specific granules of human neutrophils. The early and significant rise in serum levels of YKL40 may signal early tissue damage. The increase in YKL-40 may have been larger if the patients had not been treated with prednisolone, since glucocorticoid administration to patients with non-infectious inflammatory diseasesleads to reductions in serum YKL40 within 1 day (Julia S. Johansen, personal communication). The decline found from day 0 to day 1 for PIIINP and PINP is most probably due to the glucocorticoid treatment. The lack of significant changes in PIIINP and PINP does not rule out their importance in this context as at least PIIINP already has been shown to be elevated in ARDS (33,34). The lower numbers of blood samples for analysis of these markers may have hidden important information on the degree of fibrosis or simply reflect the low degree of exposure. Since peak levels of serum zinc appear late in the followup period and are merging peak levels of fibrinogen, zinc level in serum seem to act as an acute phase reactant rather than expressing the initial degree of exposure. Prolonged observation after exposure is in general recommended in the presence of diffuse infiltration on chest radiograph, pyrexia, leucocytosis and elevated levels of zinc in urine or blood. Such findings may indicate exposure of zinc chloride or zinc oxychloride to the alveoli. Recently, the immediate damage to lung parenchyma has also been determined by technetium-99m DTPA radioaerosol inhalation lung scintigraphy (35). Since none of the exposed soldiers from this study developed ARDS one may only speculate that the development of acute phase reactants and DLCO,,,, after ZCS exposure may
CHLORIDE
SMOKE
889
contribute to early diagnosis of ARDS. But it is important to point out that even modest exposure to ZCS leads to quantifiable changes in both lung function and acute phase reactants. This finding justifies the inclusion of phase reactants and determination of DLCO,,,, when monitoring patients after exposure to ZCS.
Acknowledgements We would like to thank Dr B. Eisner and J. Brandt for reagents (antigen and antibodies) used in the analysis of PINP.
References 1. Evans EH. Casualties following exposure to zinc chloride smoke. Lancet 1945; 2: 370. 2. Hjortso E, Qvist J, Bud MI, et al. ARDS after accidental inhalation of zinc chloride smoke. Intensive Care Med 1988; 14: 17-24. 3. Fischer H. Accidents Caused by Combat and Smoke Chemicals, and Their Sequelae. [Unf%lle durch Kampfund Nebelstoffe und ihre Folgen] German. Wehrmed. Mschr 1969; 13: 355-359. 4. Cullumbine H. The toxicity of screening smokes. JR Army Med Corps 1957; 103: 119-122. 5. The Danish Army Chief. Textbook for privates. Occupational environment and safety. [Grundbog for Menige, Arbejdsmiljti og -sikkerhed.]. 3 ed. DK-6900
Skjern: J. Strandbygaard A/S; 1994: additional page. 6. Schenker MB, Speizer FE, Taylor JO. Acute upper respiratory symptoms resulting from exposure to zinc chloride aerosol. Environ Res 1981; 25: 317-324. 7. Johnson FA, Stonehill RB. Chemical Pneumonjtis from Inhalation of Zinc Chloride. Dis Chest 1961; 40: 619-624. 8. Pedersen C, Hansen CP, Gronfeldt W. [Zinc chloride smoke poisoning following the use of smoke ammunition]. Ugeskr Laeger 1984; 146: 2397-2399. 9. Matarese SL, Matthews Jl. Zinc chloride (smoke bomb) inhalational lung injury. Chest 1986; 89: 308-309. 10. Whitaker PH. Radiological appearences of the chest following partial asphyxation by a smoke screen. BrJ Radio1 1945; 18: 396397. 11. Allen MB, Crisp A, Snook N, Page RL. ‘Smoke-bomb’ pneumonitis. Respir Med 1992; 86: 165-166. 12. Milliken JA, Waugh D, Kadish ME. Acute interstitial pulmonary fibrosis caused by a smoke bomb. Can Med Assoc J 1963; 88: 3639. 13. Homma S, Jones R, Qvist J, Zapol WM, Reid L. Pulmonary vascular lesions in the adult respiratory distress syndrome caused by inhalation of zinc chloride smoke: a morphometric study. Hum Pathol 1992; 23: 45550. 14. Lumsden RB, Weir MC. Subglottic stenosis after exposure to a high concentration of screening smoke (zinc chloride). Br Med J 1945; 1: 554555.
890
B. ZER.4HNETAL.
15. Freitag A, Caduff B. [ARDS caused by military zinc fumes exposure]. [German]. Schweiz Med Wochenschr 1996; 126: 1006-1010. 16. Wang YT, Lee LK, Poh SC. Phosgene poisoning from a smoke grenade. Eur J Respir Dis 1987; 70: 126128. 17. Kofoed-Enevoldsen A, Molvig JC, Zerahn B, Ebbehoj NE. [Zinc chloride smoke pollution. Effects of minimal exposure]. Ugeskr Laeger 1997; 159: 7318-7321. 18. Anonymous. Guidelines for the measurement of respiratory function. Recommendations of the British Thoracic Society and the Association of Respiratory Technicians and Physiologists [see comments]. Respir Med 1994; 88: 165-194. 19. Cotes JE, Dabbs JM, Elwood PC, Hall AM, McDonald A, Saunders MJ. Iron-deficiency anaemia: its effect on transfer factor for the lung (diffusiong capacity) and ventilation and cardiac frequency during sub-maximal exercise. Clin Sci (Colch) 1972; 42: 325-335. 20. Johansen JS, Jensen HS, Price PA. A new biochemical
marker for joint injury. Analysis of YKL-40 in serum and synovial fluid. Br J Rheumatol 1993; 32: 949-955. 21. Johansen JS, Moller S, Price PA, et al. Plasma YKL40: a new potential marker of fibrosis in patients with alcoholic cirrhosis? Stand J Gastroenterol 1997; 32: 582-590. 22. Orum 0, Hansen M, JensenCH, et al. Procollagen type
I N-terminal propeptide (PINP) as an indicator of type I collagen metabolism: ELISA development, reference interval, and hypovitaminosis D induced hyperparathyroidism. Bone 1996; 19: 157-163. 23. Risteli J, Niemi S, Trivedi P, Maentausta 0, Mowat AP, Risteli L. Rapid equilibrium radioimmunoassay for the amino-terminal propeptide of human type III procollagen. Clin Chem 1988; 34: 715-718. 24. Groth S, Dirksen A, Dirksen H, Rossing N. Intraindividual variation and effect of learning in lung function examinations. A population study. Bull Eur Physiopath01 Respir 1986; 22: 3542. 25. Karlsson N. Poisoning from smoke grenades is not due to phosgene [letter]. Eur Respir J 1988; 1: 575-575.
26. Brain JD, Valberg PA. Deposition of aerosol in the respiratory tract. Am Rev Respir Dis 1979; 120: 1325-1373. 27. Schmahl K. [Clinical signs in zinc-chloride smoke intoxication (author’s transl)]. Pneumonologie 1974; 150: 161-169. 28. Brown RF, Marrs TC, Rice P, Masek LC. The histopathology of rat lung following exposure to zinc oxide/hexachlorethane smoke or installation with zinc chloride followed by treatment with 70% oxygen. Environ Health Perspect 1990; 85: 81-87. 29. Helm KU, Renovanz H-D, Schmahl K, Clarmann M. Zinc Chloride Smoke Poisoning and its Treatment. Wehrmed Mschr 1971; 7: 203-217. 30. Chinet T, Jaubert F, Dusser D, Dane1 C, Chretien J, Huchon GJ. Effects of inflammation and fibrosis on pulmonary function in diffuse lung fibrosis. Thorax 1990; 45: 675-678. 31. Johansen JS, Cintin C, Jorgensen M, Kamby C, Price PA. Serum YKL-40: a new potential marker of prognosis and location of metastases of patients with recurrent breast cancer. Eur J Cancer 1995; 31A: 1437-1442. 32. Renkema GH, Boot RG, Au FL, et al. Chitotriosidase, a chitinase, and the 39-kDa human cartilage glycoprotein, a chitin-binding lectin, are homologues of family 18 glycosyl hydrolases secreted by human macrophages. Eur.J.Biochem. 1998; 251: 504-509. 33. Entzian P, Huckstadt A, Kreipe H, Barth J. Determination of serum concentrations of type III procollagen peptide in mechanically ventilated patients. Pronounced augmented concentrations in the adult respiratory distress syndrome. Am Rev Respir Dis 1990; 142: 1079-1082. 34. Farjanel J, Hartmann DJ, Guidet B, Luquel L, Offenstadt G. Four markers of collagen metabolism as possible indicators of diseasein the adult respiratory distress syndrome. Am Rev Respir Dis 1993; 147: 1091-1099. 35. Lin W-Y, Kao C-H, Wang S-J. Detection of acute inhalation injury in fire victims by means of technetium-99m DTPA radioaerosol inhalation lung scintigraphy. Eur J Nucl Med 1997; 24: 125-129.
MEDICINE
(1999) 93,885-890
Pulmonary damage after modest exposure to zinc chloride smoke B. ZERAHN*,A. KOFOED-ENEVOLDSEN~, B. V. JENSEN*, J. MQLVIG?, N. EBBEHDJ~, J. S. JOHANSEN~ AND I.-L. KANSTRUP* *Department of Clinical Physiol. and Nuclear Medicine, Herlev Hospital, University of Copenhagen, ?Department of Endocrinology, Herlev Hospital, University of Copenhagen,SDepartment of Occupational and Environmental Medicine, Bispebjerg Hospital, University of Copenhagen and §Department of Medicine, Division of Rheumatology, University of Copenhagen, Hvidovre Hospital, Copenhagen, Denmark Thirteen soldiers (11 men and two women) were exposed to zinc chloride smoke (ZCS) during a combat exercise. Even though their initial symptoms were modest, a prolonged follow up with lung function testing and blood samples was undertaken due to previous caseswith fatal outcome after exposure to ZCS. Four weeks after exposure there were statistically significant declines from baseline values in lung diffusion capacity and total lung capacity of 16.2% and 4.3%, respectively. At the same time plasma levels of fibrinogen and zinc were significantly elevated, though mainly within the normal range. All variables showed a tendency towards normalization at follow up 8 weeks and 6 months after exposure. These findings indicate an unexpected quantifiable damage to lung parenchyma with a remarkable delay after modest exposure to zinc chloride smoke despite sparse initial symptoms. Exposure to high concentrations of ZCS may induce adult respiratory distress syndrome (ARDS) after a symptom free period of up to 12 days from exposure. Even though none of the soldiers in the present study developed ARDS the assessmentof lung diffusion capacity and acute phase reactants is proposed as a supplement when monitoring patients after exposure to ZCS. RESPIR. MED.
(1999) 93, 885-890
0 -_I_---_-._-
Introduction Zinc chloride smoke (ZCS) screens are used for both military and civilian screening purposes. The use of these smoke bombs in confined spaces has been widely abandoned due to several lethal incidents (l-3). However, the use of ZCS in open air is considered safe within certain limitations (4,5). Immediate symptoms after exposure may be irritation of the eyes and the upper respiratory tract with lachrymation, metallic taste, coughing, hoarseness and soreness of the throat, which in mild cases last from a few hours to a few weeks (6,7). In more severe cases nausea vomiting, chest pain, and headache also occur (8,9). Clinical and paraclinical findings include cyanosis, pyrexia, leucocytosis (mostly polymorfonuclear leucocytes), reduced vital capacity (VC) and forced expiratory volume in 1 set (FEV1) at the peak of symptoms with recovery during convalescence, increased zinc urinary excretion and Paperreceived7 May 1999and accepted18 May 1999. Correspondence should be addressed to: Bo Zerahn, MD, Department .of Clinical Physiology and Nuclear Medicine, FrederiksbergHospital, 2000 Frederiksberg,Denmark. Fax: +45 39 77 76 22; E- mail: [email protected] 0954-6111/99/120885+06 $12*00/O
-_l(--.-.--_
1999 HARCOURT
PUBLISHERS
LTD
- -- __.-“,“_x.-..., I .,~..-‘~.--~.-I__--i-..~.--^---
.-^I”~-
various degrees of hypoxemia. Chest radiographs may show diffuse infiltration that in some cases persist in various degrees without clinical symptoms. In more severe cases emphysematous bullae and pneumothorax occur (10). Exposure to high concentrations may induce adult respiratory distress syndrome (ARDS) after a symptom free period of up to 12 days from exposure (11,12). The outcome is frequently fatal, and in such cases autopsies have shown extensive interstitial and intra-alveolar space fibrosis and arterial and venous lumen reduction as well as obliteration and widespread occlusion of microvessels. However, the exact mechanism by which ARDS is triggered is unknown (13). In exceedingly high concentrations of ZCS death is caused within hours by oedema of the larynx and spasm of the glottis (14). In lesser but still very high concentrations severe haemorrhagic ulceration and excoriation of the upper respiratory tract may lead to a fatal outcome (1). At present, only casuistic reports have dealt with diffusion capacity after ZCS exposure (11,15,16). We describe a 6-month follow up of lung function variables and specific serum variables in 13 soldiers, who were exposed to modest concentrations of ZCS during a routine military exercise. 0
1999 HARCOURT
PUBLISHERS
LTD
886
B. ZERAHN
ETAL.
Material and methods PATIENTS Thirteen soldiers, (two women and 11 men) aged 19 to 26 (median age 20 years) were exposed to ZCS during a combat exercise. Eleven of the soldiers were never smokers while the remaining two were present smokers (4.25 and 6 pack-years, respectively). The soldiers were together with nine unexposed colleagues either defending or attacking a combat exercise building with a wide door and window openings. A total of two smoke canisters were ignited in the open air, but due to wind conditions the smoke drifted into the combat exercise building through the open windows and doors. Two of the soldiers inside the building were exposed for 5-10 min without airway protection. The remaining 11 soldiers were briefly exposed while running through the plume of smoke and/or upon entering the combat exercise building. Primarily 12 of the soldiers were hospitalized on the day of exposure for observation due to soreness of the throat, coughing, chest pain and mild headache. None of the soldiers had severe respiratory symptoms when admitted, but four of them complained of dyspnoea during the combat exercise. Upon admittance all soldiers were treated with inhalation of 1 mg budesonide and 200 mg intravenous hydrocortisone followed by 75 mg oral prednisolone after 12 h. Four soldiers were then treated with oral prednisolone 40 mg daily for 1 week and then gradual reduction in dose with Smg every third day. Three of the soldiers were treated because of the duration and concentration of smoke exposition and the fourth was treated because of persistent coughing and chest pain. This soldier had a laryngoscopy 5 days after exposure, showing signs of hypopharyngeal inflammation interpreted as chemical pharyngitis. One soldier was treated for 18 h with intravenous N-acetylcysteine (140 mg kg-’ day-‘) until blood samples were obtained showing normal values for plasma zinc. The thirteenth soldier was admitted to hospital the day after exposure due to mild initial symptoms. He was not treated with steroids but joined the follow-up programme.
METHODS All soldiers had spirometry and lung diffusion capacity testing performed using a Pulmonary Function Testing System 1070 (Medical Graphics Corp. Minnesota, U.S.A.). One soldier however, had her first lung function test performed at another hospital in Copenhagen county. The results from the latter test have been used as her baseline values and have not been presented in the preliminary study on our results (17). All soldiers had the first lung function test upon the first day after exposure. The following pulmonary variables were recorded at each session: Vital capacity (VC), forced expiratory volume in 1 set (FEVt), maximal expiratory flow when xx% of the FVC remains to be delivered (MEF,), total lung capacity (TLC), lung diffusion capacity with adjustment for actual
haemoglobin content (DLCO,,,) and diffusing capacity per alveolar volume (DL/Va) were recorded (18). TLC and DLCOcorr were determined with the single breath technique after inhalation of a gas mixture containing carbon monoxide and neon with the latter as an indicator of the dilution. The better of two results that differed no more than 3% was used as the final test result. Only experienced lung function technicians performed the tests. Lung diffusion capacity was adjusted for haemoglobin level according to Cotes et al. (19). Blood samples were obtained from all soldiers on the second day and subsequently after 1, 2,4, and 8 weeks. At day 0 blood samples were obtained from 10 of the soldiers and 29 weeks after exposure from 11 of the soldiers. From these blood samples plasma levels of zinc, fibrinogen and Creactive protein were determined along with haemoglobin, blood sedimentation rate and differentiated while blood cell counts. Also serum levels of YKL40 (human cartilage glycoprotein-39) and the aminoterminal propeptides of type 1 and type III collagen (PINP and PIIINP) were determined on each blood sample. YKL40 is an 18glycosyl hydrolase that has been suggested to function in the degradation of extracellular matrix or tissue inflammation (20,21). Both PINP and PIIINP are regarded as markers of formation of collagen type I and III respectively (22,23). However, blood samples were only sufficient for determination of YKL40, PIIINP and PINP at day 0 and day 1 for 10 of the soldiers and from day 2 for 12 of the soldiers. Serum levels of YKL-40 and PIIINP were determined by radio immunoassays (20,23) and PINP by enzyme linked immunosorbent assay (22). Radiographs of the thorax and arterial blood gas analyses were also obtained from day 1 after exposure and during follow-up.
DATA ANALYSIS Changes in lung function variables and blood analyses with time were tested using the Wilcoxon non-parametric test for paired data and the Friedmann non-parametric test for several related samples.
Results Data regarding changes in lung function variables for the 13 exposed soldiers are listed in Table 1. All baseline values for lung function were within normal range when compared to a Danish reference population (24). Significant declines in DLCO,,,, TLC, and Dl/Va.were found, with the lowest values observed 4 weeks after exposure. MEFzs was also lower 4 weeks after exposure compared to baseline values, but the Friedmann test only showed a trend towards significance. The development of changes in DLCO,,, compared to initial values is illustrated in Fig. 1. A significant increase in VC was found 2 weeks after exposure. During follow-up we observed no significant drifts or jumps with time during routine calibration tests of the pulmonary test system. We were not able to demonstrate any correlation between the degree of exposure,
1. Lung function 0 (25-75%)
(4.5-8.5) (3.5-6.4) (1.3-2.0) (2.9-5.3) (6.3-l 1.6) (8.5-17.3) (2.1-3.0)
7.0 5.4 1.6 4.6 8.6 14.2 2.4
after mild exposure
median
variables
7.1 5.4 1.7 4.7 8.2 13.2 2.3
1 (25-75%)
(4.48.8) (34-6.7) (0.8-2.3) (3.0-5.6) (6.1-10.6) (8.33155)* (1.8-2.7)***
median
to zinc chloride
7.3 5.7 2.0 4.6 8.0 12.8 2.2
2 (25-75%)
(3.8-9.2) (3.5-6.7)*** (0.8-2.3) (3G6.1) (5.3-10.6)(*) (8.2-15.1)*** (1.7-2.5)***
median
smoke
6.7 5.5 1.6 4.6 7.5 11.9 2.1
4 (25575%)
(4.3-9.0)** (3.66.5) (1.G2.0) (2.9-6.0) (5.7-9.5)** (7.3-13.7)*** (1.7-2.4)***
median 7.0 5.5 1.9 4.7 8.3 12.7 2.1
median 6.8 5.6 1.8 4.5 8.2 13.7 2.2
8 (25-75%)
(44-9.0) (3.66.5) (l&2.1) (2.8-6.0) (5.7-9.9) (7.3314.9)*** (l&2.5)***
median
(4.8-8.6)* (3&6.8)* (1.3-2.1) (3.1-6.1) (6.3-10.1) (7&14.3)* (1.8-2.8)
29 (25-75%)
21.4*** 20.2*** 10.4(*) 8.6 9.5(*) 28.6*** 26.6***
Friedman test Chi-square
Significance levels are indicated by asterixes: (*)O.l > P > 0.05;* P < 0.05;** P-c 0.02; *** P
TLC(l) vc (1) RV (1) FEVl (1) ME& (l/s) DLCO,,,, (mM/min/mmHg) DL/Va (mM/min/mmHg/l)
Weeks after exposure
TABLE
888
B.
ZERAHN
ETAL.
from baseline values were found during the remaining follow-up period. For all soldiers both plasma levels of PIIINP and PINP declined from day 0 to day 1 by median 32%, PcO.01, n= 10. No significant changes with time for either of the two collagen formation markers were found hereafter. Chest radiographs and arterial blood gas analyses remained normal for all the soldiers during follow-up. * **
**
16 -
Discussion
14 12 --
-
10 86-
18 -
81
-
-
0.3
o
: IIII1IIII
.‘,
012
4 Weeks
8 after mild exposure
I
29
to zincchloridesmoke 1. Changes from values at day of DLCO,,,, in percent and of plasma zinc and fibrinogen in ,u mol-’ after modest exposure to zinc chloride smoke. *P
FIG
**p
***p
treatment with steroids, or smoking and the changes in lung function variables. None of the soldiers had leucocytosis upon admittance. But the initial levels of both neutrophils and monocytes were significantly higher than 29 weeks after exposure (36% and 25% respectively, P-c 0.01). Only the four soldiers who received oral prednisolone had levels of neutrophils above normal range during follow-up. Plasma levels of fibrinogen were significantly elevated during a period from 1-8 weeks after exposure compared to initial values and some peak values exceeded normal range (Fig. 1). Plasma levels of zinc were above initial values at 28 weeks after exposure (Fig. 1). Individual peak levels for plasma zinc were dispersed over a wide range of time from the second day to 8 weeks after exposure and remained within normal range. Five out of the 13 soldiers had slightly elevated plasma levels of CRP and eight out of 13 had a slight increase in blood sedimentation rate during the period from 2-8 weeks after exposure. We were not able to demonstrate a correlation between changes in lung function variables and changes in plasma levels of zinc or fibrinogen. Plasma level of YKL-40 increased by median 1118% from day zero to day two (n = 10, P= 0.005). No differences
We have demonstrated a significant decreasein DLCO,,,,, TLC and Dl/Va after modest open-air exposure to ZCS with a corresponding increase in serum levels of fibrinogen and zinc. These findings indicate that sufficient amounts of ZCS must have passed deeply enough into the lower respiratory tract to induce a quantifiable reduction in lung function. Preliminary data and statistics on changes in DLCO,,,, after exposure to ZCS from the present study have already been published (17). The present study is an expanded version with prolonged follow-up, improved computed statistical analyses and extended blood sample analyses. Gas transfer for carbon monoxide after exposure to ZCS has only been described previously in three singular cases. One was found normal (1 l), another had a slight reduction in DLCO 4 months after exposure (15) and the third one showed moderate to severe reduction 2 weeks after exposition with recovery after 2-3 months (16). A ZCS bomb contains hexachlorethane (C&), zinc oxide (ZnO) and calcium silicide (CaS$ along with minor amounts of aluminium, magnesium and ammonium chloride. When ignited these components form mainly zinc chloride (ZnClz), which again in aqueous solution forms hydrochloric acid (HCl) and zinc oxychloride (ZnClOH) (4). These three chemicals are all corrosive. Also minor amounts of phosgene (COCQ carbon monoxide (CO) and polyaromatic hydrocarbons are produced, but the pulmonary and toxic impact of these compounds is generally considered to be of little or no importance in this context (25). In confined spaces an ignited smoke bomb rapidly absorbs all available moisture from the air. In such a case the main component of the smoke will be zinc chloride in minute sized particles ranging from 0.01 to 25~ (7) which may pass deep into the respiratory tract (26). With increasing air humidity successivelylarger amounts of zinc chloride will be hydrolysed into hydrochloric acid and zinc oxychloride. The hydrochloric acid absorbs comparably more water and forms larger particles mainly deposited in the upper respiratory tract (27). In rats ZCS induces pulmonary oedema, alveolitis and subsequently fibrosis (28). The pathological findings after fatal cases in man are similar. They include extensive interstitial and intra-alveolar fibrosis, diffuse microvascular obliteration and widespread occlusion of the pulmonary arteries (29). The progression of ARDS is suggested to be mediated through a programmed release of mediators from recruited and activated inflammatory cells and stimulation
ZINC
of mediator cascades that include cytokine interactions (13). The increase in serum levels of fibrinogen and zinc and the decline in ,DLCO,,,, TLC, Dl/Va and to a lesserextent MEFz5 is compatible with an ulcerative and fibrotic process in the lung alveoli and smaller bronchioli with a delayed onset and an apparent maximum at 4 weeks after exposure (30). However, the modest symptoms of the soldiers did not justify attempts to verify this by histology neither by biopsy nor by bronchoalveolar lavage. The following healing response seemsslow, and one cannot rule out the possibility of permanent damage from the present data. The slight increase in VC and FVC found 2 weeks after exposure can be explained by adaptation to the test procedure. It is also noteworthy that the reduction VC and FEVr described in patients with heavy exposure to ZCS was not demonstrable in the present material. This may indicate that DLCO,,,,, TLC and Dl/Va are more sensitive procedures in such cases with mild exposure. Although a sufficiently matched control group was not obtainable due to the acute circumstances of the study, we are confident that the observed changes are not due to circadian variations or lung function technicians. YKL40 is presumed to function in tissue remodelling and serum YKL40 levels are elevated in patients with active rheumatoid arthritis (20) in patients with liver fibrosis (21) and in patients with recurrent breast cancer and metastases to bone and viscera (31). YKL-40 is secreted by human macrophages (32) and is a matrix protein in specific granules of human neutrophils. The early and significant rise in serum levels of YKL40 may signal early tissue damage. The increase in YKL-40 may have been larger if the patients had not been treated with prednisolone, since glucocorticoid administration to patients with non-infectious inflammatory diseasesleads to reductions in serum YKL40 within 1 day (Julia S. Johansen, personal communication). The decline found from day 0 to day 1 for PIIINP and PINP is most probably due to the glucocorticoid treatment. The lack of significant changes in PIIINP and PINP does not rule out their importance in this context as at least PIIINP already has been shown to be elevated in ARDS (33,34). The lower numbers of blood samples for analysis of these markers may have hidden important information on the degree of fibrosis or simply reflect the low degree of exposure. Since peak levels of serum zinc appear late in the followup period and are merging peak levels of fibrinogen, zinc level in serum seem to act as an acute phase reactant rather than expressing the initial degree of exposure. Prolonged observation after exposure is in general recommended in the presence of diffuse infiltration on chest radiograph, pyrexia, leucocytosis and elevated levels of zinc in urine or blood. Such findings may indicate exposure of zinc chloride or zinc oxychloride to the alveoli. Recently, the immediate damage to lung parenchyma has also been determined by technetium-99m DTPA radioaerosol inhalation lung scintigraphy (35). Since none of the exposed soldiers from this study developed ARDS one may only speculate that the development of acute phase reactants and DLCO,,,, after ZCS exposure may
CHLORIDE
SMOKE
889
contribute to early diagnosis of ARDS. But it is important to point out that even modest exposure to ZCS leads to quantifiable changes in both lung function and acute phase reactants. This finding justifies the inclusion of phase reactants and determination of DLCO,,,, when monitoring patients after exposure to ZCS.
Acknowledgements We would like to thank Dr B. Eisner and J. Brandt for reagents (antigen and antibodies) used in the analysis of PINP.
References 1. Evans EH. Casualties following exposure to zinc chloride smoke. Lancet 1945; 2: 370. 2. Hjortso E, Qvist J, Bud MI, et al. ARDS after accidental inhalation of zinc chloride smoke. Intensive Care Med 1988; 14: 17-24. 3. Fischer H. Accidents Caused by Combat and Smoke Chemicals, and Their Sequelae. [Unf%lle durch Kampfund Nebelstoffe und ihre Folgen] German. Wehrmed. Mschr 1969; 13: 355-359. 4. Cullumbine H. The toxicity of screening smokes. JR Army Med Corps 1957; 103: 119-122. 5. The Danish Army Chief. Textbook for privates. Occupational environment and safety. [Grundbog for Menige, Arbejdsmiljti og -sikkerhed.]. 3 ed. DK-6900
Skjern: J. Strandbygaard A/S; 1994: additional page. 6. Schenker MB, Speizer FE, Taylor JO. Acute upper respiratory symptoms resulting from exposure to zinc chloride aerosol. Environ Res 1981; 25: 317-324. 7. Johnson FA, Stonehill RB. Chemical Pneumonjtis from Inhalation of Zinc Chloride. Dis Chest 1961; 40: 619-624. 8. Pedersen C, Hansen CP, Gronfeldt W. [Zinc chloride smoke poisoning following the use of smoke ammunition]. Ugeskr Laeger 1984; 146: 2397-2399. 9. Matarese SL, Matthews Jl. Zinc chloride (smoke bomb) inhalational lung injury. Chest 1986; 89: 308-309. 10. Whitaker PH. Radiological appearences of the chest following partial asphyxation by a smoke screen. BrJ Radio1 1945; 18: 396397. 11. Allen MB, Crisp A, Snook N, Page RL. ‘Smoke-bomb’ pneumonitis. Respir Med 1992; 86: 165-166. 12. Milliken JA, Waugh D, Kadish ME. Acute interstitial pulmonary fibrosis caused by a smoke bomb. Can Med Assoc J 1963; 88: 3639. 13. Homma S, Jones R, Qvist J, Zapol WM, Reid L. Pulmonary vascular lesions in the adult respiratory distress syndrome caused by inhalation of zinc chloride smoke: a morphometric study. Hum Pathol 1992; 23: 45550. 14. Lumsden RB, Weir MC. Subglottic stenosis after exposure to a high concentration of screening smoke (zinc chloride). Br Med J 1945; 1: 554555.
890
B. ZER.4HNETAL.
15. Freitag A, Caduff B. [ARDS caused by military zinc fumes exposure]. [German]. Schweiz Med Wochenschr 1996; 126: 1006-1010. 16. Wang YT, Lee LK, Poh SC. Phosgene poisoning from a smoke grenade. Eur J Respir Dis 1987; 70: 126128. 17. Kofoed-Enevoldsen A, Molvig JC, Zerahn B, Ebbehoj NE. [Zinc chloride smoke pollution. Effects of minimal exposure]. Ugeskr Laeger 1997; 159: 7318-7321. 18. Anonymous. Guidelines for the measurement of respiratory function. Recommendations of the British Thoracic Society and the Association of Respiratory Technicians and Physiologists [see comments]. Respir Med 1994; 88: 165-194. 19. Cotes JE, Dabbs JM, Elwood PC, Hall AM, McDonald A, Saunders MJ. Iron-deficiency anaemia: its effect on transfer factor for the lung (diffusiong capacity) and ventilation and cardiac frequency during sub-maximal exercise. Clin Sci (Colch) 1972; 42: 325-335. 20. Johansen JS, Jensen HS, Price PA. A new biochemical
marker for joint injury. Analysis of YKL-40 in serum and synovial fluid. Br J Rheumatol 1993; 32: 949-955. 21. Johansen JS, Moller S, Price PA, et al. Plasma YKL40: a new potential marker of fibrosis in patients with alcoholic cirrhosis? Stand J Gastroenterol 1997; 32: 582-590. 22. Orum 0, Hansen M, JensenCH, et al. Procollagen type
I N-terminal propeptide (PINP) as an indicator of type I collagen metabolism: ELISA development, reference interval, and hypovitaminosis D induced hyperparathyroidism. Bone 1996; 19: 157-163. 23. Risteli J, Niemi S, Trivedi P, Maentausta 0, Mowat AP, Risteli L. Rapid equilibrium radioimmunoassay for the amino-terminal propeptide of human type III procollagen. Clin Chem 1988; 34: 715-718. 24. Groth S, Dirksen A, Dirksen H, Rossing N. Intraindividual variation and effect of learning in lung function examinations. A population study. Bull Eur Physiopath01 Respir 1986; 22: 3542. 25. Karlsson N. Poisoning from smoke grenades is not due to phosgene [letter]. Eur Respir J 1988; 1: 575-575.
26. Brain JD, Valberg PA. Deposition of aerosol in the respiratory tract. Am Rev Respir Dis 1979; 120: 1325-1373. 27. Schmahl K. [Clinical signs in zinc-chloride smoke intoxication (author’s transl)]. Pneumonologie 1974; 150: 161-169. 28. Brown RF, Marrs TC, Rice P, Masek LC. The histopathology of rat lung following exposure to zinc oxide/hexachlorethane smoke or installation with zinc chloride followed by treatment with 70% oxygen. Environ Health Perspect 1990; 85: 81-87. 29. Helm KU, Renovanz H-D, Schmahl K, Clarmann M. Zinc Chloride Smoke Poisoning and its Treatment. Wehrmed Mschr 1971; 7: 203-217. 30. Chinet T, Jaubert F, Dusser D, Dane1 C, Chretien J, Huchon GJ. Effects of inflammation and fibrosis on pulmonary function in diffuse lung fibrosis. Thorax 1990; 45: 675-678. 31. Johansen JS, Cintin C, Jorgensen M, Kamby C, Price PA. Serum YKL-40: a new potential marker of prognosis and location of metastases of patients with recurrent breast cancer. Eur J Cancer 1995; 31A: 1437-1442. 32. Renkema GH, Boot RG, Au FL, et al. Chitotriosidase, a chitinase, and the 39-kDa human cartilage glycoprotein, a chitin-binding lectin, are homologues of family 18 glycosyl hydrolases secreted by human macrophages. Eur.J.Biochem. 1998; 251: 504-509. 33. Entzian P, Huckstadt A, Kreipe H, Barth J. Determination of serum concentrations of type III procollagen peptide in mechanically ventilated patients. Pronounced augmented concentrations in the adult respiratory distress syndrome. Am Rev Respir Dis 1990; 142: 1079-1082. 34. Farjanel J, Hartmann DJ, Guidet B, Luquel L, Offenstadt G. Four markers of collagen metabolism as possible indicators of diseasein the adult respiratory distress syndrome. Am Rev Respir Dis 1993; 147: 1091-1099. 35. Lin W-Y, Kao C-H, Wang S-J. Detection of acute inhalation injury in fire victims by means of technetium-99m DTPA radioaerosol inhalation lung scintigraphy. Eur J Nucl Med 1997; 24: 125-129.