Mortality prognostic factors for patients with ventilator associated pneumonia

T¹p chÝ y - d−îc häc qu©n sù sè 8-2020 95 MORTALITY PROGNOSTIC FACTORS FOR PATIENTS WITH VENTILATOR ASSOCIATED PNEUMONIA Bui Van Phap1, Le Van Nam2 SUMMARY Objectives: To investigate some mortality prognostic factors in patients with ventilator associated pneumonia (VAP) at Military Hospital 103 (from July 2018 to September 2019). Subjects and methods: A retrospective and prospective study was conducted on 47 patients. 78.7% were males, mean age 57.64 ± 18.55 years, who were diagnosed with VAP according to diagnostic criteria of ATS in 2016. Results: Overall mortality rate was 29.8%. Statistically significant factors related to mortality prognosis in patients with VAP included: SOFA score > 5 with OR = 4.85 (95%CI, 1.12 - 17.03) and p 10 ng/mL with OR = 2.78 (95%CI, 0.57 - 13.51) and p < 0.05; Early onset VAP (< 5 days) with OR = 0.4 (95%CI, 0.09 - 1.63) and p < 0.05. Conclusions: SOFA scores, PCT concentration, and VAP onset were prognostic factors associated with mortality in VAP patients. * Keywords: Ventilator associated pneumonia; Mortality prognostic factors. INTRODUCTION Hospital-acquired pneumonia (HAP) is defined as pneumonia that occurs 48 hours or more after admission, which was not incubating any respiratory symptoms of infection or new or progressive lesions on pulmonary X-ray at the time of admission. Ventilator associated pneumonia is a form of HAP that appears 48 hours after endotracheal intubation or tracheostomy, which is a common complication in patients in intensive care unit with the incidence of approximately 8 - 20% in all resuscitated patients and 10% in patients on mechanical ventilation. VAP prolongs the time for mechanical ventilation from 7.6 - 11.5 days, increases the number of days of treatment from 11.5 - 13.1 days and the cost of treatment by more than 40,000 USD/patient with the death rate ranging from 20 - 50% [2]. There are many factors associated with mortality in VAP patients such as the patient's age, underlying diseases, SOFA scores, blood PCT levels, VAP onset, sepsis status, pathogens of VAP... The investigation of these factors helps to predict the severity of the disease and propose measures to reduce the mortality rate in VAP patients. Therefore, we conducted a research aiming: To investigae some prognostic factors of death in patients with ventilator associated pneumonia at Military Hospital 103 from 7/2018 - 9/2019. 1Military Hospital 175 2Military Hospital 103, Vietnam Military Medical University Corresponding author: Bui Van Phap (france0459@gmail.com) Date received: 14/8/2020 Date accepted: 05/10/2020 T¹p chÝ y - d−îc häc qu©n sù sè 8-2020 96 SUBJECTS AND METHODS 1. Subjects 47 patients were diagnosed with VAP being treated at 103 Military Hospital from June 2018 to October 2019 * Inclusion criteria: The patient was diagnosed with VAP according to the standards of ATS in 2016 [3]. - Symptoms appear 48 hours after mechanical ventilation (via endotracheal tube or through tracheostomy). - Lung X-ray: new or progressive lesion lasting more than 48 hours with 2 of the 3 following signs: + Temperature > 38.5oC or < 35oC. + WBC > 10 G/L or WBC < 4 G/L + Opaque sputum or change of sputum properties. - Positive bronchial, sputum culture. * Exclusion criteria: - AFB sputum, bronchial fluid positive. - Isolate fungi, viruses in sputum, bronchial fluid. - Lung cancer, lung abscess. 2. Methods This is a retrospective and prospective study, descriptive analysis. Patients were divided into 2 groups (survivors and death groups) and compare some factors between two groups about patient age, onset of pneumonia (early onset VAP: < 5 days, late onset VAP: ≥ 5 days), concentration PCT, SOFA score, sepsis status, etiology of VAP. All study patients were registered in the unified form. Patient’s information was based on medical records for research. * Data processing: Data were collected and processed by SPSS 20.0 software. RESULTS AND DISCUSSION 1. Treatment results Table 1: Treatment results Treatment results Number of patients Percentage % Survivors 33 70.2 Death 14 29.8 In our study, the survival rate was 70.2%; The overall death rate was 29.8%. Our results are similar to other studies. In 2018, the study by Hoang Khanh Linh on VAP in ICU, Bach Mai Hospital found that the mortality rate was 34.6%, the cure rate was 65.4% [1]. Pham Thai Dung researched on VAP at 103 Military Hospital in 2013 found that the death rate was 23.81%; the survival rate was 76.19% [2]. T¹p chÝ y - d−îc häc qu©n sù sè 8-2020 97 2. Some prognostic factors of death in patients with VAP Table 2: Relationship between age and mortality rate. Survivors (n = 33) Death (n = 14) Age n % n % p OR (95%CI) ≤ 60 18 54.55 6 42.86 > 60 15 45.45 8 57.14 > 0.05 1.6 (0.45 - 5.65) ± SD 55.21 ± 18.67 63.36 ± 17.57 > 0.05 In the mortality group, 57.14% of patients were above 60 years of age, higher than the survival group, this rate was 45.45%, the difference was not statistically significant with p > 0.5 and OR = 1.6. The average age of the survivors group was 55.21 ± 18.67 years, lower than that in the death group (63.36 ± 17.57 years), the difference was not statistically significant with p = 0.171. When comparing the age in the death and survival groups, we found that the difference was not statistically significant (p > 0.05). Thus, old age was a factor that increases the risk of death due to low resistance to bacterial infections; but it was not an independent predictor of mortality in patients with VAP. Table 3: Relationship between the onset of VAP and mortality rate. Survivors (n = 33) Death (n = 14) VAP onset n % n % p OR (95%CI) Early 6 18.2 5 35.7 Late 27 81.8 9 64.3 < 0.05 0.4 (0.09 - 1.63) The mortality rate of patients with early onset of VAP (< 5 days) was 35.7%, higher than that of the survivor, the difference was statistically significant with p < 0.05 and OR = 0.4, (95%CI, 0.09 - 1.63). Cases of early onset VAP often had worse clinical course, higher risk of death than the group with late onset VAP. Our results were similar to other authors’ findings. Uzzwal Kumar Mallick found that the mortality rate among the early VAP was 26.7%, lower than that of the late VAP group of 62.5% (p = 0.011) [4]. Mohan Ju et al. found that the mortality rate of early VAP was 51.7%, the survival of early VAP was 26.5% (p = 0.025) [5]. Table 4: Relationship between PCT and mortality rate. Survivors (n = 33) Death (n = 14) PCT n % n % p OR (95%CI) ≤ 10 ng/mL 25 75.8 9 64.3 > 10 ng/mL 4 24.2 4 35.7 < 0.05 2.78 (0.57 - 13.51) T¹p chÝ y - d−îc häc qu©n sù sè 8-2020 98 PCT is a reliable marker for diagnosing bacterial infections, especially systemic infections. PCT values > 10 ng/mL are common in severe systemic inflammatory response due to sepsis and septic shock. The mortality rate of patients with PCT concentration > 10 ng/mL accounted for 35.7%, higher than the survival group with the rate of 24.2%, the difference was statistically significant with OR = 2.78; 95%CI, 0.57 - 13.51 (p < 0.05). Pham Thai Dung's study on the value of PCT in monitoring treatment outcomes in VAP patients found that patients with good treatment results had lower PCT levels compared to other patients, patients with good results all have decreased PCT levels. Patients with PCT concentrations higher than 10 ng/mL had higher mortality rates than survivor patients. The difference was statistically significant with p < 0.05 [2]. Table 5: Relationship between SOFA scores and mortality rate. Survivors (n = 33) Death (n = 14) SOFA scores n % n % p OR (95%CI) SOFA ≥ 5 12 36.4 10 71.4 SOFA < 5 21 63.6 4 28.6 0.03 4.85 (1.12 - 17.03) ± SD 4.12 ± 0.26 6.29 ± 0.54 0.02 The average SOFA score of the survivor was 4.12 ± 0.26 points, lower than that of the mortality group (average SOFA score was 6.29 ± 0.54 points). There was statistical significance with p < 0.05. The patients with SOFA score > 5 points had a mortality of 71.4% higher than patients with SOFA score 5 is a prognostic factor of mortality in VAP patients with p < 0.05 and OR = 4.85 (95%CI, 1.12 - 17.03). Karakuzu Z, et al studied mortality prognostic factors for VAP patients who found that SOFA scores in survivors group (6 points) were lower than the death group (8 points) (p 6 had prognostic significance of mortality with a sensitivity of 73.8% and specificity 71.2%. Juthamas Inchai et al concluded that SOFA scores were a prognostic factor for mortality in VAP patients with OR = 3,4 and p < 0.001. The SOFA score at the time of diagnosis of VAP in the survival group (7 points) was higher than the death group (4 points). SOFA score > 5 was related to mortality rate with sensitivity of 87.7% and specificity of 75.4% [6]. Table 6: Relationship between sepsis status at VAP diagnosis and mortality rate. Survivors (n = 33) Death (n = 14) Sepsis status at VAP diagnosis n % n % p OR (95%CI) Yes 9 27.3 6 42.9 No 24 72.7 8 57.1 0.295 0.5 (0.13 - 1.84) T¹p chÝ y - d−îc häc qu©n sù sè 8-2020 99 In the group of survivors, the proportion of patients with sepsis at the diagnosis of VAP was 27.3%, lower than that of the death group with the rate of 42.9%. However, the difference was not statistically significant with p > 0.05. Our results were similar to other studies’ findings. Research by Aušra ˇCiginskien˙ et al showed that the sepsis status in the survivors was 35.2% lower than that in the mortality group with the rate of 64.8%, but the difference was not significant for p = 0.65. According to Ilias I. Siempos et al, sepsis status was not a prognostic factor of death in patients with VAP with OR = 3.6 and p = 0.09 [7]. Table 7: Relationship between pathogens and mortality rate. Survivors (n = 33) Death (n = 14) Disease-causing pathogens n % n % p OR (95%CI) Acinetobacter baumannii 17 51.5 7 50.0 0.924 1.06 (0.30 - 3.71) Pseudomonas aeruginosa 8 24.2 3 21.4 0.835 1.17 (0.26 - 5.28) Klebsiella pneumoniae 3 9.1 3 21.4 0.246 0.36 (0.06 - 2.09) Others 5 15.2 1 7.2 0.452 2.32 (0.24 - 21.92) In both groups, A.baumannii was the leading cause of VAP (50% in the mortality group and 51.5% in the survival one), followed by P.aeruginosa (21.4% vs. 24.2%). This difference was not statistically significant between two groups with p > 0.05. The cause of VAP was not a prognostic factor for mortality in VAP patients. Our study was similar to the results of other studies. Kaweesak Chittawatanarat et al in a study on the prognostic factors of VAP mortality found differences in pathogens among survivors (32.7% were due to A. baumannii, 17.3% due to K .pneumoniae, 20.2% for P. aeruginosa) and death group (52.2% for A. baumannii, 17.4% for K. pneumoniae, 8.7% for P. aeruginosa) (p < 0.05 ). However, the etiology of VAP was not a predictive factor of mortality in VAP patients. Mohamed H. Afifi's study of prognostic factors in VAP patients found that the survivor group was isolated 53.3% as K. pneumoniae, 15.7% were P. aeruginosa, 10% were A. baumannii; the mortality group was isolated 25% as K. pneumoniae, 25% as P. aeruginosa, 20% as Enterobacter. Differences in pathogens between the two groups were not statistically significant. The etiology was not a prognostic factor for death in patients with VAP [8]. T¹p chÝ y - d−îc häc qu©n sù sè 8-2020 100 CONCLUSION Through a study on 47 patients diagnosed VAP at Military Hospital 103 from June 2018 to October 2019, we drew some conclusions: - The overall death rate of the study group was 29.8%. - Some prognostic factors of death in VAP patients: + SOFA score > 5 with OR = 4.85 (95%CI, 1.12 - 17.03) and p < 0.05. + PCT blood concentration > 10 ng/mL with OR = 2.78 (95%CI, 0.57 - 13.51) and p < 0.05). + Early onset VAP with OR = 0.4 (95%CI, 0.09 - 1.63) and p < 0.05. REFERENCES 1. Hoàng Khánh Linh. Nghiên cứu đặc điểm viêm phổi liên quan thở máy tại Khoa Hồi sức Tích cực Bệnh viện Bạch Mai giai đoạn 2017 - 2018. Luận văn Bác sĩ Nội trú. Bệnh viện, Đại học Y Hà Nội 2018. 2. Phạm Thái Dũng, Đỗ Quyết. Nghiên cứu đặc điểm vi khuẩn ở bệnh nhân viêm phổi thở máy. Tạp chí Y học TP. Hồ Chí Minh 2013; 17(3):159-1634. 3. Kalil Andre C, Metersky Mark L, Klompas Michael, et al. Management of adults with hospital-acquired and ventilator-associated pneumonia: 2016 clinical practice guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clinical Infectious Diseases 2016; 63(5):e61-e111. 4. Mallick Uzzwal Kumar, Faruq Mohammad Omar, Ahsan ASM Areef, et al. Spectrum of early onset and late onset ventilator associated pneumonia (vap) in a tertiary care hospital of bangladesh: A prospective cohort study. Bangladesh Critical Care Journal, 2015; 3(1):9-13. 5. Ju Mohan, Hou Dongni, Chen Shu, et al. Risk factors for mortality in ICU patients with Acinetobacter baumannii ventilator-associated pneumonia: Impact of bacterial cytotoxicity, Journal of Thoracic Disease 2018; 10(5):2608. 6. Inchai Juthamas, Pothirat Chaicharn, Liwsrisakun Chalerm, et al. Ventilator- associated pneumonia: Epidemiology and prognostic indicators of 30-day mortality. Japanese Journal of Infectious Diseases 2015; 68(3):181-186. 7. Siempos Ilias I, Vardakas Konstantinos Z, Kyriakopoulos Christos E, et al. Predictors of mortality in adult patients with ventilator- associated pneumonia: A meta-analysis. Shock 2010; 33(6):590-601. 8. Afifi Mohamed H, Elhendy Amera A, Eltaweel Moataz M, et al. Biomarker predictors of survival in patients with ventilator-associated pneumonia. Menoufia Medical Journal 2015; 28(1):254. Số đặc biệt Chào mừng Kỷ niệm 65 năm Ngày Truyền thống Bộ môn - Khoa Truyền nhiễm, Bệnh viện Quân y 103 - Học viện Quân y (20/2/1956 - 20/2/2021)