Revision of an infected total hip arthroplasty: the need for the adjustment of risk in bundled payment models for revision arthroplasty.

AIMS: The aim of this study was to investigate the differences in 30-day outcomes between patients undergoing revision for an infected total hip arthroplasty (THA) compared with an aseptic revision THA. PATIENTS AND METHODS: This was a retrospective review of prospectively collected data from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database, between 2012 and 2017, using Current Procedural Terminology (CPT) codes for patients undergoing a revision THA (27134, 27137, 27138). International Classification of Diseases Ninth Revision/Tenth Revision (ICD-9-CM, ICD-10-CM) diagnosis codes for infection of an implant or device were used to identify patients undergoing an infected revision THA. CPT-27132 coupled with ICD-9-CM/ICD-10-CM codes for infection were used to identify patients undergoing a two-stage revision. A total of 13 556 patients were included; 1606 (11.8%) underwent a revision THA due to infection and there were 11 951 (88.2%) aseptic revisions. RESULTS: Patients undergoing an infected revision had a significantly greater length of stay of more than three days (p < 0.001), higher odds of any 30-day complication (p < 0.001), readmission within 30 days (p < 0.001), 30-day reoperations (p < 0.001), and discharge to a destination other than the patient's home (p < 0.001). CONCLUSION: The findings suggest the need for enhanced risk adjustment based on the indication of revision THA prior to setting prices in bundled payment models of total joint arthroplasty. This risk adjustment should be used to reduce the chance of financial disincentives in clinical practice. Cite this article: Bone Joint J 2019;101-B:547-551.

Effect of intravenous vasopressor on spread of spinal anaesthesia and fetal acid-base equilibrium.

BACKGROUND: We previously found rostral spread of spinal plain levobupivacaine to be less with prophylactic i.v. phenylephrine than with ephedrine during Caesarean delivery. This study investigated whether rostral spread of spinal hyperbaric bupivacaine is also less with phenylephrine than with ephedrine. METHODS: The study was randomized and double blind. It compared phenylephrine 100 microg ml-1 (phenylephrine group, n=27), and ephedrine 4.5 mg ml-1 (ephedrine group, n=27), given by infusion during spinal anaesthesia for Caesarean delivery. Block height was assessed to cold and light touch sensation at 15, 30, 60, and 90-min after the spinal injection of 2.8 ml of hyperbaric 0.5% w/v bupivacaine, combined with 0.4 ml diamorphine (1 mg ml-1). Umbilical blood gas values were monitored during the study. RESULTS: Block height was similar for both groups at all of the assessment times. Umbilical artery pH was higher with phenylephrine [median 7.32 (IQR 7.28-7.34)] than with ephedrine [7.20 (7.10-7.28)] (P<0.0001). There was a strong negative correlation between umbilical artery pH and spinal-delivery interval, but only with ephedrine: phenylephrine group, r2=0.09 (P=0.17), and ephedrine group, r2=0.53 (P<0.0001). Five-minute Apgar scores were higher with phenylephrine [10 (9-10)] than ephedrine [9 (9-9)] (P=0.009). CONCLUSIONS: In contrast to its effect on spinal plain levobupivacaine, we did not find rostral spread of spinal hyperbaric bupivacaine to be less with prophylactic phenylephrine than with ephedrine. We observed an unexpectedly high incidence of fetal acidosis with ephedrine and found evidence that longer spinal-delivery intervals increase the risk of fetal acidosis developing with ephedrine, but not phenylephrine.