Pneumonia in the Pregnant Patient: A Synopsis

, and , Division of Pulmonary and Critical Care Medicine, Winthrop University Hospital, Mineola, NY


Medscape General Medicine. 1999;1(3) 

In This Article


Review of the literature shows a widely varying incidence of pneumonia in the pregnant population, reflecting both the population studied and the era in which the survey was conducted. Hopwood, Benedetti, Berkowitz, and Madinger have compiled leading surveys of pneumonia epidemiology. In 1965 Hopwood[8] studied 2720 pregnancies, finding an incidence of pneumonia of 1 in 118 deliveries. Benedetti and coworkers[9] retrospectively studied 89,219 patients in both university and county hospital settings during the 3-year period from 1972-1975, finding an incidence of pneumonia of 1 in 2288 deliveries -- a significant decline compared to that reported by Hopwood. Madinger and colleagues[10] found that pneumonia complicated 1 in 1287 deliveries at a community hospital over a 5-year period from 1983 to 1988. Finally, Berkowitz and LaSala[3] retrospectively examined 1120 case records at a large city hospital from 1988 to 1989 and found antepartum pneumonia to occur in 1 of every 367 deliveries.

This fairly recent rise in the incidence of pneumonia reported by the Madinger and Berkowitz series relative to Benedetti may be a reflection of the increasing frequency of chronic illness among pregnant women today. Perinatological advances allow us to carry a more chronically ill woman through to the completion of pregnancy, whereas previously this may not have been the case. There is also an increased incidence of newer immune deficiencies, including HIV infection, all of which may predispose the pregnant patient to other infections such as pneumonia. Other factors that place the pregnant patient at increased risk for pneumonia are tobacco use, illicit drug abuse, and prior heart disease. Admittedly, these factors present a similar additional risk to the nonpregnant patient as well, but the underlying host "immunologic defect" in the pregnant woman -- which will be discussed later -- is yet another hazard that increases the likelihood that any pneumonia developing in the pregnant woman will be more clinically significant than in her nonpregnant counterpart.

Why did Hopwood's paper demonstrate such a high incidence of pneumonia in comparison to other data collections? Perhaps his findings are due to the use of less stringent criteria for diagnosing pneumonia, resulting, then, in an overestimation of the true incidence. Reviewing the Hopwood paper more carefully does reveal a number of inadequacies. The methodology of microbiological data collection is unclear. The paper gives a list of what the authors defined as etiological agents, but does not specify the precise microbiological techniques used -- moreover the source is not identified (ie, blood versus sputum). Sputum cultures represent an area of intense controversy regarding the interpretation of data collected. The screening of sputum for quality of the specimen is also not mentioned and so we cannot be certain that some of the etiologic agents were not colonizing organisms -- an important point, because the investigators reported that all but one of their patients with pneumonia at discharge were found to be cigarette smokers (who are likely to be chronically colonized). Also of note is that Hopwood does not clarify if the patients received prior antibiotics as outpatients.

Although he reviewed over 2000 pregnancies, only 23 patients were diagnosed with pneumonia and thus the study sample size was small -- perhaps too small to make a fair estimate of the problem at that time.

In the way of commentary on data reported in the Benedetti[9] series, while these investigators clearly defined the clinical and microbiological criteria used for diagnosis, again the majority of patients (18/39) had culture-negative pneumonia. No comment was made regarding initiation of antibiotic therapy prior to admission.

In the retrospective review conducted by Madinger and colleagues[10] of medical records covering the 63-month period from 1983 to 1988 at Cedars-Sinai Medical Center, patients were identified by a computerized search using discharge diagnosis as the criterion for inclusion.

These investigators defined pneumonia clearly: a pulmonary process consistent with pneumonia, an infiltrate on chest radiograph, and absence of any other confounding infection. Madinger and associates also outlined the microbiological criteria employed. However, in reporting their results, they do not separate data according to source of organism -- this may have been important in assessing outcome (ie, how did bacteremic patients fare in comparison to nonbacteremic patients?). These investigators were aggressive in seeking out the inciting pathogens -- they cultured blood and sputum, used virology and histological, cytological, and bronchoscopic methods, and at times serology tools as well. Despite the above approach, no pathogenic organism was identified in 44% of the samples, thus leading to an underestimation of the incidence of certain etiological agents.

This is a recurring theme -- ie, all studies examining etiologic agents in community-acquired pneumonia find no pathogen in most instances (ATS Pneumonia CAP guidelines[11]. Last, Madinger and coworkers also noted that because serology was not performed on the 3 suspected cases of Mycoplasma pneumoniae, the true incidence of atypical infection was likely underestimated. Overall, Madinger's study is much more open to interpretation than Hopwood's.

As a final note in the way of critical commentary, Berkowitz and LaSala, [3] in a retrospective chart review of 1120 antepartum admissions, clearly reported the microbiological techniques employed and the organisms isolated (which are likely to be the etiologic agents and not just colonizing entities). The qualities of sputum specimens were also outlined.

Overall, then, it is evident that the incidence of pneumonia has varied widely in all the studies reviewed here -- comparing them critically presents a challenge. Pneumonia in the pregnant patient is an entity that most hospitalists, internists, and pulmonologists as well as obstetricians must be comfortable in managing.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.