Sputum Retention

ce

Table 1-Hemodynamic Data

co

WP CVP BP PVR SVR

Before equalization

After equalization

2.8Umin 19mm Hg 24cm HtO 8OI5Omm Hg 450 dynercm& 2200 dynesecm5

4.1 UOlin 13mm Hg 16cm H 20 130/70 mm Hg 400 dyn esecm 5 960 dynesecm5

co = cardiac output; WP = wedge pressure; CVP = central venous

pressure; PVR = pulmonar vascular resistance; SVR = systemic vascular resistance

A device for equalization of pleural pressure with the PEEP level setted into ventilator (10 cm HtO) was built by connecting an empty bottle to each pleural drain and a PEEP valve to the open port of the bottles. The patients cardiovascular function improved markedly, urinary output increased and, after 60 min of ventilation using the device, arterial blood gas measurements showed an impressing improvement: Peo. decreased to 40 mm Hg, POt increased to 130 mm Hg and pH to 7.44. 1llble 1 summarizes the hemodynamic data with the device and before its use. During the next eight days the level of PEEP in the ventilator and chest drains was gradually reduced and, at the 12th day, extubation was performed. Three days later, the chest tubes were removed without any further deterioration in the patients clinical condition. He was discharged home fully recovered 40 days after admission. Large bronchopleural fistulas are life-threatening conditions, and bilateral become much more distressing. Management by methods such as independent lung ventilation,' HFPP~4 etc, is generally unsuccessful. Segmental bronchial occlusion with a balloon-tipped catheter, e metal plug' or tissue glue8 may lead to bilateral lung atelectasis, increased venous admixture and risk of infection. The use of continuous suction can further increase the air leak through the fistula since, with this technique, we raise transpulmonary pressure (airway pressure minus intrapleural pressure), which is the major determinant of bronchopleural fistula flow. For this reason, continuous suction should be avoided. 9 ,IO Furthermore, attaching suction to the open port of the chest bottles can cause an additional increase in air leak and loss of a large portion of the delivered tidal volume." Equalization between PEEP and pleural pressure as described herein was highly effective and promptly improved oxygenation, Co. clearance and hemodynamic parameters. Although this method was first used by Downs in the mid 1970s for treatment of unilateral bronchopleural fistula,11 we were not able to find any case of bilateral bronchopleural fistula treated by this manner in the English literature.

Natan Weksler. M. D.; and Leon Ovadia, M.D., Department ofAnesthesiology. Hillel Yaffe MeJ1wrial Hospital. Hadera. Israel

REFERENCES 1 Aldeba SM, Hansen-Flaschen JH, Taylor E, et aI. Evaluation of high frequency jet ventilation in patients with bronchopleural fistulas by quantitation of the airleak. Anesthesiology 1985; 63:551-54 2 Sjostrand UH, Smith B, Hoff BH, et aI. Conventional and high frequency ventilation in dogs with bronchopleural fistula. Crit Care Med 1985; 13:191-93 3 Weksler N, Ovadia L. High frequency ventilation with the bennet MA 1 for respiratory support in patients with septic ARDS. J Cardioth Anesth 1988; 2:406-08

4 Dodds Hillman KM. Management of massive air leak with asynchronous independent lung ventilation. Intensive Care Moo 1982; 8:287-90 5 Carlon GC, Ray C Jr, Klain M, McCormack JR. High frequency positive pressure ventilation in management of a patient with bronchopleural6stula. Anesthesiology 1980; 52:160-62 6 Cant WF, Tinker }H, Tarhan S. Bronchial blockade in a child with a bronchopleural cutaneous fistula using a balloon-tipped catheter. Anesth Analg 1976; 55:874-75 7 Ratliff JL, Hill JD, Tucker H, et aI. Endobronchial control of bronchopleural fistula. Chest 1977; 71:98-99 8 Hartmann W, Raush U. A new therapeutic application of the fiberoptic bronchoscope. Chest 1977; 71:237 9 Powner OJ, Grenvik A. Ventilatory management of life-threatening bronchopleural fistula. A summary. Crit Care Med 1981; 9:54-58 10 Powner OJ, Cline 0, Rodnlan GH. Effects of chest tube suction on gas flow through a bronchopleural fistula. Crit Care Med 1985; 13:99-101 11 Downs )B, Chapman RL. Treatlnent of bronchopleural fistula during continuous positive pressure ventilation. Chest 1976; 69:363-66

Reprint requests: Dr. ~kslet; Depamnent of Anesthesiology, Hillel Yaffe Memorial Hospital, Hadera. Israel 38100

Long-term Antlarrhythmla Therapy To the Editor: In the August, 1988 issue, we described a patient with refractory sustained ventricular tachycardia and ventricular fibrillation who responded only to oral bretylium therapy (Chest 1988; 94:430-32). At the time of acceptance of the article he had done well for a total of nine months. However, at the end of the eleventh month of therapy the patient developed progressive anorexia and weight loss. This was felt to be due to the oral bretylium therapy; the patient was admitted to the hospital and oral bretylium discontinued. He was monitored continuously for a period of one week, during which time he had virtually no spontaneous ectopy, nor was any provoked during exercise tolerance test. The patient declined electrophysiologic testing and has done well over the past several months on no antiarrhythmic therapy. This illustrates that the arrhythmic substrate can change over time and suggests that even seemingly intractable arrhythmias associated with acute myocardial infarction may resolve and antiarrhythmic agents discontinued in selected individuals.

Blair ~ Grubb. M.D.. Assistant Professor of Medicine, Division of Cardiology, Medical College of Ohio, Toledo

Sputum Retention To the Editor: I believe that sputum retention is being overlooked in large numbers of patients with COPD and pneumonia, and that this contributes to the substantial morbidity and nlortality in these illnesses. Retained sputum is difficult to diagnose on auscultation. Sputum plugging usually occurs without atelectasis on chest x-ray film because atelectasis is a late phenomenon. Hypoxemia is ascribed to the underlying pulmonary disease. COPD patients have difficulty clearing sputum because of narrowed airways and inlpaired mucociliary clearance. This sputum becomes more tenacious as it remains uncleared. Older, dehydrated or debilitated patients with pneumonia may have difficulty raising thick sputum. Intubated patients may have plugs beyond the reach of the suction catheter. Routine CHEST I 95 I 4 I APRIL, 1989

939

pulmonary toilet is often ineffective in clearing patients of these molasses-like secretions. Sputum retention should be considered in these presentations: COPD with frequent exacerbations requiring intermittent or chronic steroid therapy, slowly resolvin!: pneumonia (particularly associated with debilitation), COPD or surgery, and intubated patients with thick secretions, disproportionate hypoxemia or those difficult to wean from the ventilator. The fiberoptic bronchoscope and remove these offers a low morbidity procedure to dia~ose sputum plugs. Peter Stangel, M.D. Pomona. NY

The Jackknife Technique in Statistical Analysis To the Editor: Dr. Prall referred to the "procedural error" in our table 4 in his editorial <:omment in the November issue of Chest (1988: 94:90304). Dr. Pralls comment is disappointin!:. predominantly because it rellel'ts a lack of understanding regardin!: the jackknife technique which we used in our study. If Dr. Prall had read our article thorou!:hly. he would have understood that this technique is designed to circumvent the precise "error" which he refers to. Modern statistical techniques such as the jackknife provide a means for obtaining an unbiased estimate ofthe prospective accuracy of a mathematical rule if applied to a <:ompletely different sample of patients. The jackknife technique is essentially the "leavin!: one man out" method. In the case of our study with a total sample of 83. the discriminant function was computed on only 82 cases. The person left out is then classified on the basis of the function rule derived for the 82 cases. This is repeated until every person in the sample has been left out once and classified. The technique presents a way of using the same sample for both rule derivation as well as classification (accuracy estimation). It provides a good. unbiased estimate of the accuracy which could be expected in a completely new sample. yet it maximizes existing patient resources. Thus the jackknifed estimate presented in table 4 and discussed in our paper provides an unbiased estimate of accuracy. In fact, this technique is applicable for a wide range of research situations. I urge Dr. Prall and any interested readers to consult the foll",'in!: biblio!:raphy for more information. AI/en Rothpearl. M.D .. New lark BIBLIOGRAPHY

Diaconis P, Efron B. Computer-intensive metbods in statistics (in particular pa!:e 127). Scientific American 1983; 116 Efron B. The jackknife, the bootstrap, and other resampling plans. SIAM monograph No. 38, Society for Industrial and Applied Mathematics, 1982 Efron B. Computers and the theory of statistics: thinking the unthinkable. SIAM Review 1979; 21:460 Lachenbruch PA, Goldstein M. Discriminant analysis. Biometrics 1979; 35:69

Diminished Translucency, a Common Roentgenographic Feature of Tropical Pulmonary Eosinophilia To the Editor: Diminished translucency is a common. less reco~ized roent!:enographic feature of tropical pulmonary eosinophilia (TPE) which

940

FIGl:RE. Chest radiograph showing characteristic diminished translucency in mid- and lower zones (bilateral) in a patient with TPE. was observed in 17 of our 25 (68 percent) patients. The majority of these patients had symptoms of four weeks duration or less. This feature was most conspicuous in mid- and lower zones, extending from the periphery towards the paracardiac regions (Fig) and cleared fastest with diethylcarbamazine therapy (within three days). I Long standing cases of TPE progress to crippling pulmonary fibrosis and insufficiency with poorer response to diethylcarbamais a necessity. zine therapy;' therefore early dia~osis DISCUSSION

We found the roentgenographic feature of diminished translucency-when present-is useful for early reco~ition of coexistent miliary tuberculosis, since this feature clears within three days. This feature is also useful when peripheral eosinophilia is interrupted due to intercurrent infection, or when typical clinical features are lacking. Webb was the first to note a "faint generalized loss oftranslucency" in 22 of 40 cases ofTPE.J Herlinger ' noted diminished translucency as the cardinal feature ofTPE in 62 percent of his cases. The exact pathologic basis of this roentgenographic feature is difficult to explain. Udwadia and Joshi,' in their pathologic studies of TPE. described "histiocytic infiltration with exudative features" in three of their patients with symptomatology of less than three weeks duration. This at least partly, along with tissue hypersensitivity during acute eosinophilic infiltrations. may contribute to the greyness of the lung fields in its early phase. A larger study with roentgenographic pathologic correlation might give beller insight into the cause and dia~ostic usefulness of this feature. K. Satish Chandra, M.D.; Sundaresh Peri, M.Sc.; Lakshrnana Moorthy. R.. M.D., and Birbhadra Roo. P.v., M.D .. Warangal, India REFERENCES

Herlinger H. Pulmonary changes in tropical eosinophilia. Br J Radiology 1963; 36:889-901 2 Udwadia FE. Tropical eosinophilia-a <:orrelation of clinical, histopathologic and lung function studies. Dis Chest 1967; 52:53138 3 Webb JKG. Diseases of the respiratory sytem. In: Trowell HC, Jelliffe DB, ed. Diseases ofchildren in the subtropics and tropics. London: Edward Arnold Ltd, 1958; 223-29 4 Udwadia FE, Joshi vv. A study of tropical eosinophilia. Thorax 1964; 19:548-54 Communications to the Editor