Abstracts From the Literature-Pulmonary Selected by David H. lngbar

DRG Payment for Long-Term Ventilator Patients-Revis-

ited. Douglass PS. Bone RC, Rosen RL. Chest 93~629, 1988.

This article provides a follow-up to an evaluation originahy presented in Chesr of the financial impact of diagnosis related group (DRG) payment for long-term ventilator-dependent Medicare patients at Rush-Presbyterian-St. Luke’s Medical Center. Since the results of our original study were presented, the Health Care Financing Administration (HCFA) has created two new DRGs for patients who have respiratory principal diagnoses to help recognize the resource intensive- ness associated with mechanical ventilator support. The original 95 patients’ payment, which was originally calcu- lated to be $2.2 million below costs, was recalculated to be $1.9 million below costs, representing a 13 percent reduction in the loss. We conclude that although HCFA’s recent remedial action is a step in the right direction, it provides little relief from the DRG system’s financial bias against long-term ventilator-dependent patients, because the new ventilator DRGs encompass only a small segment of these patients. As an alternative approach, we recommend a single DRG for patients who, regardless of their principal diag- noses, experience chronic respiratory failure requiring a minimum of three days of continuous ventilator treatment. (Reprinted with permission.)

Initial Experience With a Central Respiratory Monitoring Unit as a Cost-Saving Alternative to the Intensive Care Unit

for Medicare Patients Who Require Long-Term Ventilator Support. Krieger BP. Ershowsky P, Spivack D, Thorstenson J, Sackner MA. Chest 93:395, 1988.

The cost of delivering health care to Medicare patients in intensive care units (ICU), especially those on long-term ventilator support, has become a major financial burden to American hospitals. We established a central station, respira- tory, noninvasive monitoring unit (NIMU) on a cardiac telemetry floor where nurse to patient ratio was 1:4 to 1:6 as a cost-effective alternative to the ICU which has a ratio of 1:2. During the first five months of operation, 55 patients were admitted to this unit, of which 11 were long-term ventilator Medicare patients. Seven of these 11 patients were success- fully weaned and discharged from the hospital. Overall, a minimum of $21,724 was saved over five months by shifting these long-term ventilator patients from the ICU to the NIMU while quality of medical care remained or exceeded accepted standards. We conclude that alternative health care delivery systems such as the NIMU need to be investigated as safe, cost-reducing substitutes for selected ICU patients to prevent loss of quality medical care in the financial backlash of the present Medicare prospective payment schedule. (Re- printed with permission.)

Prognostic Factors and Life Expectancy of Patients With Acquired lmmunodeficiency Syndrome and Pneumocystis

carinii Pneumonia. Brenner M, Ognibene FP, Lack EE, Simmons JT, Suffredini AF, Lane, HC, Fauci AS, Parrillo, JE. Shelhamer JH, Masur H. Am Rev Respir Dis 136:1199, 1987.

To assess determinants of prognosis for 43 patients with the acquired immunodeliciency syndrome (AIDS) and Pneu- mocystis carinii pneumonia, objective clinical and histo- pathologic characteristics were analyzed for acute and long- term prognostic significance. Severe abnormalities on initial chest radiographs and alveolar-arterial oxygen differences (AaPO,) greater than 30 mm Hg were associated with higher mortality during the period of treatment for the acute episode (p < 0.05). Decreased long-term survival after the diagnosis of Pneumocystis pneumonia correlated with the severity of interstitial edema (a component of diffuse alveolar damage) on initial transbronchial biopsy and elevation of AaPO, at the time of diagnosis (Cox proportional hazards analysis, p < 0.05). The persistance of Pneumocystis cysts after 3 wk of therapy was associated with significantly decreased long- term survival (p < 0.05) when follow-up biopsy was per- formed in 27 of the patients. Patients with a diagnosis of Pneumocystis pneumonia before July 1985 had more advanced disease at the time of diagnosis and a worse prognosis than did those in whom the diagnosis was made after July 1985 (p < 0.05). This study demonstrates that important prognostic information can be derived from infor- mation obtained at initial presentation and follow-up bronchoscopic evaluation in patients with AIDS and Pneu- mocystis carinii pneumonia, and suggests that early detec- tion and initiation of therapy may improve chances for survival. (Reprinted with permission.)

Survival and Prognostic Factors in Severe Pneumocystis carinii Pneumonia Requiring Mechanical Ventilation. El-

Sadr W, Simberkog MS. Am Rev Respir Dis 137:1264, 1988.

Severe Pneumocystis carinii pneumonia (PCP) necessitat- ing mechanical ventilation has been associated with a high mortality rate in several previous studies. This has discour- aged physicians from recommending, as well as patients from accepting, mechanical ventilation when respiratory failure developed. Analysis of 19 records of patients with first episode PCP who were intubated and received mechanical ventilation showed a mortality of 57.8%. A constellation of clinical and laboratory findings was found that identified those patients more likely to survive, including a shorter duration of symptoms prior to admission, better arterial oxygenation on admission, deterioration of respiratory status soon after bronchoscopy, decrease in serum lactic acid dehy- drogenase and rapid improvement in arterial blood gas determinations after institution of mechanical ventilation.

Journal of Critical Care, Vol 3, No 4 (December), 1988: pp 275-284 275

276 ABSTRACTS

Mean survival for the group that recovered from PCP after mechanical ventilation was not different from that of other patients with PCP who did not require mechanical ventila- tion. Thus, patients at risk for PCP should be advised to present to medical facilities as soon as symptoms of this disease develop and, when appropriate, should be encouraged to undergo intubation with respiratory support if it becomes necessary. (Reprinted with permission.)

Mechanical Ventilation After Marrow Transplantation: Risk Factors and Clinical Outcome. Crawford S W, Schwartz DA, Petersen FB, Clark JG. Am Rev Respir Dis 137:682, 1988.

The risk factors for and the outcome of mechanical ventila- tory support after marrow transplantation were analyzed in 1,089 consecutive marrow recipients. Of 232 patients receiv- ing ventilatory support at this center, 62 (27%) survived an initial episode of ventilatory support. In contrast, 74% of nonventilated patients survived at least 6 months after trans- plantation. Among patients with malignant disease (n = 992). a multivariate regression model revealed 3 pre- transplant factors that were associated with ventilatory sup- port: age ~21 yr (RR = 1.6, 95% confidence intervals [CI] = 1.3, 2.0), hematologic malignancy in relapse (RR = 2.1, 95% CI = 1.7, 2.4), and HLA nonidentical donor-recipient pair (nonidentical graft) (RR = 1.7, 95% CI = 1.4, 2.1). Survival after ventilatory support was not found to be associated with any pretransplant or post- transplant factors examined. We conclude that although the risk factors for mechanical ventilatory support after marrow transplantation can be defined, we are unable to identify characteristics predictive of survival after such support. The implications of these findings in relation to patient selection and future research activities are discussed. (Reprinted with permission.)

Outcome of Respiratory Failure in Hematologic Malignancy. Peters SG, Meadows JA, III, Gracey DR. Chest 94:99, 1988. t

We reviewed records of patients with hematologic malig- nancy requiring mechanical ventilation (MV) from 1976 to 1985 (excluding postoperative MV ~48 hours). There were 119 episodes in 116 patients. In-hospital mortality was 82 percent. Of 21 (I 8 percent) episodes survived, median dura- tion of survival was 12 months. Survivors did not differ from nonsurvivors in age, leukocyte count, or duration of MV. Survival for chronic lymphocytic leukemia was 42 percent, for other leukemias 16 percent, Hodgkin’s disease 29 percent, and non-Hodgkin’s lymphomas, 6 percent. Bronchoscopy was performed in 28 patients, resulting in a diagnosis of infection, hemorrhage, or malignancy in I9 cases. Open lung biopsy (OLB) was obtained in 23 patients, yielding a diagnosis of interstitial inflammation or fibrosis (13 cases), drug effect (three), malignancy (two), hemorrhage (one), Pneumocystis (seven), aspergillosis (two), and Legionella (one). Only two patients survived following OLB. Despite intensive manage- ment and adequate diagnosis, respiratory failure in patients with hematologic malignancy carries a high mortality. Although these data may help identify groups with a limited

prognosis for long-term recovery, patient care must be indi- vidualized. (Reprinted with permission.)

Work of Breathing and Airway Occlusion Pressure During Assist-Mode Mechanical Ventilation. Sassoon CSH, Ma- hutfe CK, Te TT. Simmons DH. Light RW. Chest 931571, 1988.

We determined the effect of varying ventilator tidal vol- ume (VT) and inspiratory flow (ir) on the inspiratory muscle work (WI) during assist-mode mechanical ventilation (AMV) in four healthy subjects. In another four subjects, under constant chemoreceptor input, we determined the responses of neuromuscular output as assessed by the mouth occlusion pressure (P,,,) to alteration in WJ. During AMV, the inspiratory external work of breathing is partitioned between WI and ventilator work. With a constant ventilator trigger sensitivity, we calculated WI (joules/L of volume) as the difference between the area subtended by the airway pressure-inspiratory volume curves and the ordinate of the assisted breaths subtracted from that of the controlled breaths at ventilator v of 40,60 and 80 L/min and ventilator VT of 100, 125 and 150 percent spontaneous breathing VT. At all ventilator settings, WI was less than inspiratory muscle work of spontaneous breathing (SB) and was a function of both ventilator VT and -ir (p < 0.05), but ventilator v has more effect on WI. Under isocapnia and hyperoxia, we measured PO, and WI during AMV at ventilator VT of 125 percent of spontaneous breathing VT and ventilator v of 60, 80 and 100 L/min. End-expiratory lung volume remained constant. PO., during AMV was similar to that of the SB. Although WI decreased with increasing ventilator i’, PO,, did not decrease significantly. We conclude that during AMV, both ventilator v and to a lesser extent ventilator VT determine W. In healthy subjects changes in WI do not affect P, ,. (Reprinted with permission.)

Cardiovascular Effects of Conventional Positive Pressure Ventilation and Airway Pressure Release Ventilation.

Riisiinen J, Downs JB. Stock MC. Chest 93:911, 1988.

The hemodynamic sequelae of conventional positive pres- sure ventilation (CPPV), airway pressure release ventilation (APRV), and spontaneous breathing were compared with continuous positive airway pressure (CPAP) in ten anesthe- tized dogs who had ventilatory failure with and without parenchymal lung injury. The APRV corrected respiratory acidosis without significantly effecting arterial blood oxygen- ation, venous admixture, cardiovascular function, or tissue oxygen utilization. Application of CPPV precipitated marked depressions in blood pressure, stroke volume, and cardiac output. A concomitant decrease in venous admixture did not compensate for these adverse cardiovascular effects. Deterio- ration of tissue oxygen delivery resulted in oxygen supply- demand imbalance during CPPV. The results of this experi- mental study indicate that if ventilatory augmentation of subjects who require CPAP is desired, APRV will enhance alveolar ventilation without compromising circulatory func- tion and tissue oxygen balance, whereas CPPV will impair cardiovascular function significantly. (Reprinted with per- mission.)