Reduced length of stay following elective knee and hip arthroplasty is an obvious target for cost savings, but is it safe? The answer is a cautious “yes,” for the right patient. The authors of this study do a commendable job of tackling the question via use of the National Surgical Quality Improvement Program (NSQIP) database. Specifically, they divided primary arthroplasties performed among NSQIP-participating hospitals in 2011 and 2012 on the basis of length of stay, with “early” discharge defined as discharge 0 to 2 days postoperatively and “standard” discharge, 3 to 4 days postoperatively. The authors then carefully compared 30-day major-complication and readmission rates. From 2011 to 2012, 31% of the primary hip arthroplasties and 21% of the primary knee arthroplasties at NSQIP hospitals fell into the “early” category, and after controlling for demographic and clinical differences between the groups (age, ASA [American Society of Anesthesiologists] classification, etc.), complications and readmissions were no worse in the early discharge group; in fact, early discharge after total hip arthroplasty was actually associated with a small but statistically significant reduction in complication rates.
These are encouraging results and suggest that a standard stay of 3 to 4 days is often not necessary for the average patient scheduled for primary total knee or total hip arthroplasty. What the study does not support is indiscriminate use of an early discharge pathway for everyone. As the topic of length of stay and its relationship to complication and readmission rates continues to be revisited in the literature1,2, it has become increasingly apparent that patient and perioperative factors rather than length of stay are the primary driving force behind most early complications and readmissions. As Sutton et al. appropriately point out, well-known risk factors for early complications, such as tobacco use and increased body mass index (BMI), still apply, regardless of length of hospital stay. Similar findings were identified by Zmistowski et al.; medically complex patients had higher readmission rates, and shorter length of stay did not increase readmission rates3. A major caveat to the apparently neutral effect of early discharge on complication and readmission rates on appropriately risk-stratified patients is the case of the patient who develops a new hospital-acquired condition in the early postoperative period. There is particular danger for early readmission of the patient who develops a new hospital-acquired condition, as Raines et al. pointed out in a recent study using data from the Veterans Affairs Surgical Quality Improvement Program (VASQIP), and this accounted for 42% of all readmissions in their data set4. I would argue that providers should immediately reassess the discharge plan (the timing of both discharge and the first office follow-up) for any patient with a new hospital-acquired condition after total knee or total hip arthroplasty and have a low threshold for delaying discharge.
Sutton et al. noted that, even after controlling for length of stay and medical comorbidities, there was a higher 30-day readmission rate among patients who were discharged to an extended care facility (ECF) rather than to home and a higher complication rate among total knee arthroplasty patients discharged to an ECF. Zmistowski et al. saw a similar phenomenon with increased readmission rates after ECF placement3. I find this troubling. This may be due to a low threshold among ECFs for sending patients to a hospital for evaluation because of legal concerns for a missed diagnosis or delayed treatment of a complication. Perhaps some aspect of patient health or risk of complication that is not fully captured by the NSQIP or other databases leads to preferential placement of some higher-risk patients in an ECF. Regardless, as readmission rates, whether fair or not, become a surrogate measure of quality of care, I suspect that we will see more work in the next few years aimed at better understanding and reducing readmissions from ECFs.
In conclusion, early discharge within 2 days after primary knee or hip arthroplasty does not confer higher risk of an early complication or readmission in appropriately selected patients. Patients with a high complication risk due to preoperative factors, such as increased BMI, or intraoperative factors, such as prolonged operative time or increased blood loss, remain higher risk, regardless of length of stay. Early discharge plans should not supplant good clinical judgment, and a low threshold should be used to delay discharge or modify follow-up plans for patients who develop a new hospital-acquired condition in the early postoperative period.
↵* The author indicated that no external funding was received for any aspect of this work. The Disclosure of Potential Conflicts of Interest form is provided with the online version of the article.
- Copyright © 2016 by The Journal of Bone and Joint Surgery, Incorporated