Background: Hospital Acquired Pneumonia (HAP) is the most common nosocomial infection contributing to death (1). There are no accepted clinical criteria for diagnosis or prognostic markers. The recognition and treatment of HAP can be difficult, mortality rates are significant and it increases hospital stay by 7-9 days (2).
Aims: Review management of HAP in our trust. Identify adverse prognostic markers associated with increased mortality and prolonged hospital stay.
Methods: All patients with a diagnosis of HAP over a 12 month period from December 2007 were identified using the hospital coding records. A retrospective case note review was undertaken and data such as age,
sex, diagnosis on admission, co-morbidities, date of new symptoms, length of hospital stay and mortality were collected. Laboratory tests comprising white cell count (WCC), albumin and creatinine within 24 hours of onset of new symptoms were recorded We used Odds ratio and Chi square tests which were used to study the effects of laboratory parameters and medical co-morbidities on mortality and length of stay (LOS).
We identified 53 patients with HAP. Median (range) age was 80 (41-97) years and median (range) LOS following diagnosis of HAP was 5 (1-21) days. The mortality rate was 41.5 % (22/53 patients). We found that recent surgery was the only factor that had significant effect on mortality (Odds ratio 10.66, 95%CI 2.12-53.6, p=0.04). WCC, creatinine and albumin were associated with increased mortality but this was not statistically significant. However, WCC > 11 did prolong LOS (Chi squared test, p<0.05).
Conclusions: In our study, we have found that surgical patients had higher mortality when they developed HAP. The odds of dying were higher in patients with high WCC, abnormal creatinine and low albumin although in our cohort this did not attain statistical significance. In addition, high WCC was also associated with longer LOS. We believe larger studies are needed to corroborate our findings and assess the significance of other risk factors. This would help with risk stratification of patients with HAP and enable objective assessment of management options.
Available in fulltext at ProQuest