Development and Impact of an Institutional Enhanced Recovery Program on Opioid Use, Length of Stay, and Hospital Costs Within an Academic Medical Center: A Cohort Analysis of 7774 Patients.

BACKGROUND: Enhanced Recovery (ER) is a change management framework in which a multidisciplinary team of stakeholders utilizes evidence-based medicine to protocolize all aspects of a surgical care to allow more rapid return of function. While service-specific reports of ER adoption are common, institutional-wide adoption is complex, and reports of institution-wide ER adoption are lacking in the United States. We hypothesized that ER principles were generalizable across an institution and could be implemented across a multitude of surgical disciplines with improvements in length of stay, opioid consumption, and cost of care. METHODS: Following the establishment of a formal institutional ER program, ER was adopted in 9 distinct surgical subspecialties over 5 years at an academic medical center. We compared length of stay, opioid consumption, and total cost of care in all surgical subspecialties as a function of time using a segmented regression/interrupted time series statistical model. RESULTS: There were 7774 patients among 9 distinct surgical populations including 2155 patients in the pre-ER cohort and 5619 patients in the post-ER cohort. The introduction of an ER protocol was associated with several significant changes: a reduction in length of stay in 5 of 9 specialties; reduction in opioid consumption in 8 specialties; no change or reduction in maximum patient-reported pain scores; and reduction or no change in hospital costs in all specialties. The ER program was associated with an aggregate increase in profit over the study period. CONCLUSIONS: Institution-wide efforts to adopt ER can generate significant improvements in patient care, opioid consumption, hospital capacity, and profitability within a large academic medical center.

Enhanced recovery program for minimally invasive and vaginal urogynecologic surgery.

INTRODUCTION AND HYPOTHESIS: Enhanced recovery protocols (ERPs) are evidenced-based interventions designed to standardize perioperative care and expedite recovery to baseline functional status after surgery. There remains a paucity of data addressing the effect of ERPs on pelvic reconstructive surgery patients. METHODS: An ERP was implemented at our institution including: patient counseling, carbohydrate loading, avoidance of opioids, goal-directed fluid resuscitation, immediate postoperative feeding and early ambulation. Patients undergoing elective pelvic reconstructive surgery before and after implementation of the ERP were identified in this cohort study. RESULTS: One hundred eighteen patients underwent pelvic reconstructive surgery within the ERP compared with 76 historic controls. Reductions were seen in length of hospital stay (29.9 vs. 27.9 h, p = 0.04), total morphine equivalents (37.4 vs. 19.4 mg, p < 0.01) and total intravenous fluids administered (2.7 l vs. 1.5 l, p < 0.0001). Hospital discharges before noon doubled (32.9 vs. 60.2%, p < 0.01). More patients in the ERP group ambulated on the day of surgery (17.1 vs. 73.7%, p < 0.01) and ambulated at least two times the day following surgery (34.2 vs. 72.9%, p < 0.01). No differences were seen in average pain scores (highest pain score 7.39 vs. 7.37, p = 0.95), hospital readmissions (3.9 vs. 3.4%, p = 0.84), or postoperative complications (6.58 vs. 8.47%, p = 0.79). Patient satisfaction significantly improved. ERP was not associated with an increase in 30-day total hospital costs. CONCLUSIONS: Implementation of ERP for pelvic reconstructive surgery patients was associated with a reduced length of hospital stay, improved patient satisfaction, and decreased administration of intravenous fluids and opioids without an increase in complications, readmissions, or hospital costs.

A pilot study of patient-centered outcome assessment using PROMIS for patients undergoing colorectal surgery.

PURPOSE: Few studies have assessed patient-reported outcomes following colorectal surgery. The absence of this information makes it difficult to inform patients about the near-term effects of surgery, beyond outcomes assessed by traditional clinical measures. This study was designed to provide information about the effects of colorectal surgery on physical, mental, and social well-being outcomes. METHODS: The NIH Patient-Reported Outcomes Measurement Information System (PROMIS®) Assessment Center was used to collect patient responses prior to surgery and at their routine postoperative visit. Four domains were selected based on patient consultation and clinical experience: depression, pain interference, ability to participate in social roles and activities, and interest in sexual activity. Multilevel random coefficient models were used to assess the change in scores during the follow-up period and to assess the statistical significance of differences in trends over time associated with key clinical measures. RESULTS: In total, 142 patients were consented, with 107 patients completing pre- and postoperative assessments (75%). Preoperative assessments were typically completed 1 month prior to surgery (mean 29.5 days before, SD = 19.7) and postoperative assessments 1 month after surgery (mean 30.7 days after, SD = 9.2), with a mean of 60.3 days between assessment dates. Patients demonstrated no statistically significant changes in scores for pain interference (-0.18 points, p = 0.80) or the ability to participate in social roles and activities (0.44 points, p = 0.55), but had significant decreases in depression scores between pre- and postoperative assessments (-1.6 points, p = 0.03) and near significant increases in scores for interest in sex (1.5 points, p = 0.06). Pain interference scores for patients with neoadjuvant chemotherapy significantly increased (3.5 points, p = 0.03). Scores for the interest in sex domain decreased (worsened) for patients with oncologic etiology (-3.7 points, p = 0.03). No other differences in score trends by patient characteristics were large enough to be statistically significant at the p < 0.05 threshold. CONCLUSION: These data suggest that the majority of patients quickly return to baseline physical, mental, and social function following colorectal surgery. This information can be used preoperatively to counsel patients about the typical impact of colorectal surgery on quality of life.

Enhanced Recovery Implementation in Major Gynecologic Surgeries: Effect of Care Standardization.

OBJECTIVE: To examine implementing an enhanced recovery after surgery (ERAS) protocol for women undergoing major gynecologic surgery at an academic institution and compare surgical outcomes before and after implementation. METHODS: Two ERAS protocols were developed: a full pathway using regional anesthesia for open procedures and a light pathway without regional anesthesia for vaginal and minimally invasive procedures. Enhanced recovery after surgery pathways included extensive preoperative counseling, carbohydrate loading and oral fluids before surgery, multimodal analgesia with avoidance of intravenous opioids, intraoperative goal-directed fluid resuscitation, and immediate postoperative feeding and ambulation. A before-and-after study design was used to compare clinical outcomes, costs, and patient satisfaction. Complications and risk-adjusted length of stay were drawn from the American College of Surgeons' National Surgical Quality Improvement Program database. RESULTS: On the ERAS full protocol, 136 patients were compared with 211 historical controls and the median length of stay was reduced (2.0 compared with 3.0 days; P=.007) despite an increase in National Surgical Quality Improvement Program-predicted length of stay (2.5 compared with 2.0 days; P=.009). Reductions were seen in median intraoperative morphine equivalents (0.3 compared with 12.7 mg; P<.001), intraoperative (285 compared with 1,250 mL; P<.001) and total intravenous fluids (-917.5 compared with 1,410 mL; P<.001), immediate postoperative pain scores (3.7 compared with 5.0; P<.001), and total complications (21.3% compared with 40.2%; P=.004). On the ERAS light protocol, 249 patients were compared with 324 historical controls and demonstrated decreased intraoperative and postoperative morphine equivalents (0.0 compared with 13.0 mg; P<.001 and 15.0 compared with 23.6 mg; P<.001) and decreased intraoperative and overall net intravenous fluids (P<.001). Patient satisfaction scores showed a marked and significant improvement on focus questions regarding pain control, nurses keeping patients informed, and staff teamwork; 30-day total hospital costs were significantly decreased in both ERAS groups. CONCLUSION: Implementation of ERAS protocols in gynecologic surgery was associated with a substantial decrease in intravenous fluids and morphine administration coupled with reduction in length of stay for open procedures combined with improved patient satisfaction and decreased hospital costs.

Standardization of care: impact of an enhanced recovery protocol on length of stay, complications, and direct costs after colorectal surgery.

BACKGROUND: Colorectal surgery is associated with considerable morbidity and prolonged length of stay (LOS). Recognizing the need for improvement, we implemented an enhanced recovery (ER) protocol for all patients undergoing elective colorectal surgery at an academic institution. STUDY DESIGN: A multidisciplinary team implemented an ER protocol based on: preoperative counseling with active patient participation, carbohydrate loading, multimodal analgesia with avoidance of intravenous opioids, intraoperative goal-directed fluid resuscitation, immediate postoperative feeding, and ambulation. Discharge requirements remained identical throughout. A before and after study design was undertaken comparing patients before (August 2012 to February 2013) and after implementation of an ER protocol (August 2013 to February 2014). Risk stratification was performed using the NSQIP risk calculator to calculate the predicted LOS for each patient based on 23 variables. RESULTS: One hundred and nine consecutive patients underwent surgery within the ER protocol compared with 98 consecutive historical controls (conventional). The risk-adjusted predicted LOS was similar for each group at 5.1 and 5.2 days. Substantial reductions were seen in LOS, morphine equivalents, intravenous fluids, return of bowel function, and overall complications with the ER group. There was a $7,129/patient reduction in direct cost, corresponding to a cost savings of $777,061 in the ER group. Patient satisfaction as measured by Press Ganey improved considerably during the study period. CONCLUSIONS: Implementation of an ER protocol led to improved patient satisfaction and substantial reduction in LOS, complication rates, and costs for patients undergoing both open and laparoscopic colorectal surgery. These data demonstrate that small investments in the perioperative environment can lead to large returns.

Cardiac output monitoring: a contemporary assessment and review.

OBJECTIVE: An increasing number of minimally or noninvasive devices are available to measure cardiac output in the critical care setting. This article reviews the underlying physical principles of these devices in addition to examining both animal and human comparative studies in an effort to allow clinicians to make informed decisions when selecting a device to measure cardiac output. DATA SOURCES: Peer-reviewed manuscripts indexed in PubMed. STUDY SELECTION: A systematic search of the PubMed database for articles describing the use of cardiac output monitors yielded 1,526 sources that were included in the analysis. DATA EXTRACTION: From all published cardiac output monitoring studies reviewed, the animal model, number of independent measurements, and correlation between techniques was extracted. DATA SYNTHESIS: Comparative studies in animals and humans between devices designed for measurement of cardiac output and experimental reference standards indicate thermodilution and Doppler-based techniques to have acceptable accuracy across a wide range of hemodynamic conditions, with bioimpedance techniques being less accurate. Thermodilution devices are marginally more accurate than Doppler-based devices but suffer from slower response time, increased invasiveness, and require stable core temperatures, good operator technique, and a competent tricuspid valve. Doppler-based techniques are less invasive and offer beat-to-beat measurements and excellent trending ability, but are dependent on accurate beam alignment and knowledge of aortic cross-sectional area. Studies of newer devices, such as pulse contour analysis, partial rebreathing, and pulse wave velocity, are far less in number and are primarily based on comparisons with thermodilution-based cardiac output measurements. Studies show widely ranging results. CONCLUSION: Thermodilution is relatively accurate for cardiac output measurements in both animals and humans when compared to experimental reference standards. Doppler-based techniques appear to have similar accuracy as thermodilution pulmonary artery catheters. Bioimpedance, pulse contour, partial rebreathing, and pulse wave velocity-based devices have not been studied as rigorously; however, the majority of studies included in this analysis point towards decreased accuracy.

Assessment of the agreement between photoplethysmographic and arterial waveform respiratory variation in patients undergoing spine surgery.

Respiratory variation in the arterial blood pressure and photoplethysmographic (PPG) waveforms have both been shown to predict the haemodynamic response to volume administration. Whether or not the two can be considered interchangeable is controversial. Twenty-three patients undergoing spine surgery received both a 20 gauge intra-arterial catheter and a Masimo adult adhesive SpHb sensor connected to a Radical-7 monitor. Pulse pressure variation (PPV) was calculated off-line at 1-min intervals. Pleth Variability Index (PVI) and Perfusion Index data were recorded. After exclusion of outliers, agreement between PPV and PVI was assessed using a repeated measures Bland-Altman approach. Concordance between changes in PPV and PVI was assessed using a four-quadrant plot with a 20% zone of exclusion. In total, 6549 min of data were collected. Repeated measures Bland-Altman analysis identified a bias of 2.2% and 95% confidence intervals of ±15.3% (limits of agreement -13.1 and +17.6%). The concordance rate between changes in PPV and changes in PVI was 51%. The agreement between respiratory variation in the arterial blood pressure and PPG waveforms is poor and these two should not be considered interchangeable. Changes in PPV are unrelated to changes in PVI. The data, combined with recently published work from other authors, suggests that the low frequency oscillations in the PPG waveform are not related to the low frequency oscillation in the systemic arterial blood pressure tracing and may be related to changes in venous pressure, peripheral tone or other physiologic phenomena yet to be described.