CLEVELAND CLINIC JOURNAL OF MEDICINE VOLUME 74 • E-SUPPLEMENT 1 SEPTEMBER 2007 S7 ANITHA RAJAMANICKAM, MD Department of Hospital Medicine, Cleveland Clinic, Cleveland, OH PREETHI PATEL, MD Department of Hospital Medicine, Cleveland Clinic, Cleveland, OH ALI USMANI, MD Department of Hospital Medicine, Cleveland Clinic Clinical Assistant Professor of Medicine, Cleveland Clinic Lerner College of Medicine, Cleveland, OH Q: Are routine preoperative chest radiographs necessary in asymptomatic patients undergoing noncardiothoracic surgery? A: Routine preoperative chest radiographs (chest x-rays [CXRs]) are not indicated for asympto- matic patients younger than age 50 years who do not have risk factors for postoperative pulmonary compli- cations (PPCs). Patients with significant risk factors for PPCs may warrant a preoperative CXR irrespec- tive of age. For asymptomatic patients older than 50 years with no risk factors, there is insufficient evi- dence for or against ordering routine CXRs. Pulmonary complications: Just as common as cardiac complications The American College of Physicians’ (ACP) 2006 guidelines on preoperative pulmonary risk stratifica- tion for noncardiothoracic surgery notably state that PPCs are just as prevalent as cardiac complications. 1,2 The risk for PPCs increases with age and with patient- and procedure-related risk factors. 1,2 Risk factors for PPCs that may warrant a preoperative CXR include the following: 1–3 • American Society of Anesthesiologists physical status classification of II or greater • Functional dependency • Known cardiopulmonary disease • Upper abdominal or thoracic procedures, or sur- gery for an abdominal aortic aneurysm. Few studies evaluate postoperative pulmonary complications The practice of routinely obtaining preoperative CXRs originated during World War II to detect tuber- culosis infection, now a rarity in developed nations. 4 Today clinicians often order CXRs as part of a routine preoperative evaluation in order to screen for abnor- malities that may affect surgical risk or outcomes. Few studies that have assessed abnormal versus normal preoperative CXRs have used PPCs as the pri- mary end point. A systematic review conducted to support the 2006 ACP guidelines found that only four such studies (two univariate analyses and two multi- variate analyses) evaluated this outcome. Only two of these studies showed that an abnormal CXR was a statistically significant predictor of PPCs. 1 The first of these studies, by Lawrence et al, 5 found that both abnormal lung examination and an abnor- mal CXR were statistically significant predictors of PPCs, but these authors did not indicate if patients with abnormal CXRs also had abnormal physical exam findings. Also, the cohort population consisted predominantly of male veterans with a high preva- lence of smoking and chronic obstructive pulmonary disease. 5 The second study, by Bluman et al, 6 which was designed to evaluate the effects of smoking on PPCs, showed that an abnormal CXR was a predictor of PPCs. Abnormal CXRs have little effect on management Most studies looking at the utility of preoperative CXRs have evaluated changes in surgery date or anes- thesia management as the primary end point. In one meta-analysis of 21 studies assessing the value of pre- operative CXRs, only 10% of routine preoperative CXRs were abnormal, and in only 1.3% (95% confi- dence interval [CI], 0 to 2.8%) were these abnormali- ties unanticipated after a thorough history and physi- cal exam. 7 Furthermore, the CXR findings changed management in only 0.1% of the patients (95% CI, 0 to 0.6%). In a Canadian study of 1,000 patients who had a preoperative CXR, 74 had abnormalities on CXR, and 68 of these patients (92%) had a history or symp- toms of cardiorespiratory disease. 8 Surgery was delayed in 1.3% of the patients with radiographic abnormalities. However, without symptoms or perti- nent medical history, abnormal CXRs did not predict a worse clinical outcome. In a United Kingdom study by the Royal College of Radiologists involving 10,619 patients undergo- ing nonacute, noncardiopulmonary surgery, preoper- All authors reported that they have no commercial affiliations or financial interests that pose a potential conflict of interest with this article. IMPACT CONSULTS