[Research on risk factors of short-term outcome in AIDS patients with pneumocystis pneumonia]

Objective: To investigate the prognostic risk factors of acquired immunodeficiency syndrome (AIDS) patients with pneumocystis pneumonia (PCP), and to establish risk models for predicting early outcome. Methods: The clinical data of 418 AIDS patients with PCP admitted to Department of Infectious Diseases, Beijing You'an Hospital, Capital Medical University from January 2008 to May 2016 were retrospectively analyzed.The patients were divided into death group and survival group according to clinical outcome during hospitalization.Data of the two groups were collected including general information and laboratory test results.Multivariate Logistic regression was used to analyze risk factors affecting prognosis of patients, establish prognostic models and evaluate predictive value of the model. Results: Of the 418 AIDS patients with PCP, 388 cases were male and 30 cases were female, aged from 5 to 82 years, mean age was (40±12) years.There were 82 patients in the death group and 336 patients in the survival group.Disease course, bacterial infection and alveolar-arterial oxygen pressure difference(P(A-a)O(2)), serum lactate dehydrogenase(LDH), white blood cell (WBC), neutrophil (N), alanine aminotransferase (AST), urea nitrogen (BUN) and serum potassium (K) were significantly higher in the death group than those in the survival group (all P<0.05), and arterial oxygen pressure (PaO(2)), blood oxygen saturation (SpO(2)), CD4(+) T lymphocyte count, lymphocyte (L) , hemoglobin (Hb), platelet (PLT), albumin (ALB), prealbumin (PALB), cholinesterase (CHE), cholesterol (CHO), serum chlorine (Cl) and serum sodium (Na) were significantly lower in the death group than those in the survival group (all P<0.05). Multivariate Logistic regression analysis showed that P(A-a)O(2, )ALB, LDH, N and CD4(+) T lymphocyte count were prognostic factors of AIDS complicated with PCP.Prognostic index=9.736+ 0.112×P(A-a)O(2)-0.719×ALB+ 0.006×LDH+ 0.355×N-0.021×CD4.ROC curve of the short-term prognostic model was 0.985 (95%CI 0.977-0.994), with P value 0.000, cut-off value 0.907, sensitivity 92.0% and specificity 98.8%.The mortality rate increased with the increase of equation value. Conclusions: P(A-a)O(2, )ALB, LDH, N and CD4(+) T lymphocyte count are independent risk factors to predict short-term prognosis in these patients.The short-term prognostic model based on independent risk factors is useful in guiding clinical treatment.