Pulmonary embolism (PE) is a common complication after orthopedic surgery. As symptoms are variable and may be absent, decisions regarding which post‐surgical patients to evaluate with CT pulmonary angiography (CTPA) may not be straightforward. Clinical prediction tools, such as the Wells score, have not been validated for risk stratification in this population. D‐dimer is likewise of limited utility after surgery. The purpose of this study is to explore the relationship between the presence of PE on CTPA with readily available patient characteristics in a post‐operative orthopedic population.
We performed a retrospective cohort study. 331 inpatient records at an orthopedic specialty hospital identified as containing CTPA results were reviewed. Vital signs, lab values, and other clinical data were evaluated for an association with the presence or absence of PE on the index CTPA. Chi‐square, t‐test and Mann‐Whitney tests were used for univariate analysis. All predictor variables found to have associations with p‐values < 0.10 were dichotomized and re‐evaluated by multivariate logistic regression.
Of 331 patients who underwent CTPA, 52 (15.7%) were diagnosed with PE. Clinical factors identified as predicting for or against PE are listed in Table 1, along with corresponding p‐values and odds ratios.
The clinical factors showing the strongest association with PE on CTPA were female gender, low‐grade tachycardia (peak heart rate 100‐109 during the preceding 48 hours), hypoxia (SpO2 < 88% either immediately prior to the scan or in the preceding 48 hours), positive serum troponin, and status‐post total knee replacement. The clinical factors associated with absence of PE were age < 50, higher grade tachycardia (peak HR > 110 in the preceding 48 hours), use of general anesthesia, and status‐post hip repair or spine surgery. BNP < 250 showed a trend towards protecting against PE, but BNP levels were unavailable in 96 (29.0%) patients. Notably, blood pressure, BMI, and prior diagnosis of either malignancy or heart failure showed no relationship with CTPA results.
On multivariate analysis, only status‐post total knee replacement and low grade tachycardia remained significant predictors of PE, with adjusted odds ratios of 2.40 and 2.395, respectively.
In post‐operative orthopedic patients, readily available clinical data, such as vital signs and lab results, may be used to guide decisions regarding whom to scan for PE with CTPA. However, because no combination of clinical criteria could be developed to effectively rule in or rule out PE, diagnostic decisions should remain individualized.
To cite this abstract:Baghdadi J, Iturrate E. Risk Factors for Pulmonary Embolism in Orthopedic Patients Undergoing Post‐Operative Ct Pulmonary Angiography. Abstract published at Hospital Medicine 2014, March 24-27, Las Vegas, Nev. Abstract 95. Journal of Hospital Medicine. 2014; 9 (suppl 2). https://www.shmabstracts.com/abstract/risk-factors-for-pulmonary-embolism-in-orthopedic-patients-undergoing-postoperative-ct-pulmonary-angiography/. Accessed May 26, 2019.