Cardiac Surgery in Patients With Opioid Use Disorder: An Analysis of 1.7 Million Surgeries

Management of Patients With Tricuspid Valve Endocarditis and Ongoing Intravenous Drug Abuse

Right-sided infective endocarditis in patients with intravenous drug abuse portends a worse prognosis. Data on optimal management strategy in this situation are scarce. We describe outcomes of 2 different treatment strategies, including a patient treated conservatively with favorable intermediate-term results and another who was treated surgically and developed recurrent endocarditis. (Level of Difficulty: Intermediate.).

Impact of preexisting opioid dependence on morbidity, length of stay, and inpatient cost of urological oncological surgery

OBJECTIVES

To determine the incidence of preexisting opioid dependence in patients undergoing elective urological oncological surgery. In addition, to quantify the impact of preexisting opioid dependence on outcomes and cost of common urologic oncological procedures at a national level in the USA.

METHODS

We used the National Inpatient Sample (NIS) to study 1,609,948 admissions for elective partial/radical nephrectomy, radical prostatectomy, and cystectomy procedures. Trends of preexisting opioid dependence were studied over 2003-2014. We use multivariable-adjusted analysis to compare opioid-dependent patients to those without opioid dependence (reference group) in terms of outcomes, namely major complications, length of stay (LOS), and total cost.

RESULTS

The incidence of opioid dependence steadily increased from 0.6 per 1000 patients in 2003 to 2 per 1000 in 2014. Opioid-dependent patients had a significantly higher rate of major complications (18 vs 10%; p < 0.001) and longer LOS (4 days (IQR 2-7) vs 2 days (IQR 1-4); p < 0.001), when compared to the non-opioid-dependent counterparts. Opioid dependence also increased the overall cost by 48% (adjusted median cost $18,290 [IQR 12,549-27,715] vs. $12,383 [IQR 9225-17,494] in non-opioid-dependent, p < 0.001). Multivariable analysis confirmed the independent association of preexisting opioid dependence with major complications, length of stay in 4th quartile, and total cost in 4th quartile.

CONCLUSIONS

The incidence of preexisting opioid dependence before elective urological oncology is increasing and is associated with adverse outcomes after surgery. There is a need to further understand the challenges associated with opioid dependence before surgery and identify and optimize these patients to improve outcomes.

Prevalence and Outcomes of Opioid Use Disorder in Pediatric Surgical Patients: A Retrospective Cohort Study

BACKGROUND

Chronic opioid use among adolescents is a leading preventable public health problem in the United States. Consequently, a sizable proportion of surgical patients in this age group may have a comorbid opioid use disorder (OUD). No previously published study has examined the prevalence of OUD and its impact on postoperative morbidity or mortality in the adolescent surgical population. Our objective was to investigate the prevalence of comorbid OUD and its association with surgical outcomes in a US adolescent surgical population. We hypothesized that OUD among adolescent surgical patients is on an upward trajectory and that the presence of OUD is associated with higher risk of postoperative morbidity or mortality.

METHODS

Using the pediatric health information system, we performed a 1:1 propensity score-matched, retrospective cohort study of adolescents (10-18 years of age) undergoing inpatient surgery between 2004 and 2019. The primary outcome was inpatient mortality. The secondary outcomes were surgical complications and postoperative infection. We also evaluated indicators of resource utilization, including mechanical ventilation, intensive care unit (ICU) admission, and postoperative length of stay (LOS).

RESULTS

Of 589,098 adolescents, 563 (0.1%) were diagnosed with comorbid OUD (563 were matched on OUD). The prevalence of OUD in adolescents undergoing surgery increased from 0.4 per 1000 cases in 2004 to 1.6 per 1000 cases in 2019, representing an average annual percent change (AAPC) of 9.7% (95% confidence interval [CI], 5.7-13.9; P value < .001). The overall postoperative mortality rate was 0.50% (n = 2941). On univariable analysis, mortality rate was significantly higher in adolescents with comorbid OUD than those without comorbid OUD (3.37% vs 0.50%; P < .001). Among propensity-matched pairs, comorbid OUD diagnosis was associated with an estimated 57% relative increase in the risk of surgical complications (adjusted relative risk [aRR], 1.57; 95% CI, 1.24-2.00; P < .001). The relative risk of postoperative infection was 2-fold higher in adolescents with comorbid OUD than in those without OUD (aRR, 2.02; 95% CI, 1.62-2.51; P < .001). Adolescents with comorbid OUD had an increased risk of ICU admission, mechanical ventilation, and extended postoperative LOS.

CONCLUSIONS

OUD is becoming increasingly prevalent in adolescents presenting for surgery. Comorbid OUD is an important determinant of surgical complications, postoperative infection, and resource utilization, underscoring the need to consider OUD as a critical, independent risk factor for postsurgical morbidity.

Preexisting opioid use disorder is associated with poor outcomes in hospitalized acute pancreatitis patients

INTRODUCTION

Although opioids are widely used for pain management in acute pancreatitis, the impact of opioid use disorder (OUD) on outcomes in patients with acute pancreatitis remains unknown. In the current study, we aimed to evaluate the impact of the OUD on outcomes in patients hospitalized with acute pancreatitis and delineate the trends associated with OUD and acute pancreatitis using a nationally representative sample.

METHODS

This is a retrospective cohort study of patients with acute pancreatitis using the combined releases of the year 2005-2014 of the National (Nationwide) Inpatient Sample (NIS) database. Patients over the age of 18 years with a principal diagnosis of acute pancreatitis were divided into cohorts of patients with opioid use disorders and those without. The primary measured outcome was in-hospital mortality and secondary outcomes were healthcare utilization measures, including length of stay (LOS) and hospitalization costs.

RESULTS

A total of 2 593 831 hospitalizations of acute pancreatitis were included; of which, 37 849 (1.46%) had a secondary diagnosis of OUD. Total acute pancreatitis-related hospitalizations increased from 237 882 in 2005 to 274 006 in 2014. At the same time prevalence of OUD in acute pancreatitis patients also increased from 1 to 2.1%. Patients with OUD had significantly increased mortality as compared to patients without OUD (aOR: 1.4; P < 0.001). At the same time, acute pancreatitis patients with OUD were associated with 1.3 days longer LOS as compared to other acute pancreatitis patients (P < 0.001]. The mean adjusted difference in total hospitalization costs was $2353 (P < 0.001).

CONCLUSION

OUD is associated with a significant increase in LOS, healthcare utilization cost and in-hospital mortality in patients admitted for acute pancreatitis. Therefore, clinicians should exercise caution in prescribing opioid medications to this high-risk patient population and other modalities such as nonopioid pain medications should be tried as alternatives to opioid analgesics.

Comparison of Hospitalization Trends and Outcomes in Acute Myocardial Infarction Patients With Versus Without Opioid Use Disorder

Discrepancies in medical care are well known to adversely affect patients with opioid abuse disorders (OUD), including management and outcomes of acute myocardial infarction (AMI) in patients with OUD. We used the National Inpatient Sample was queried from January 2006 to September 2015 to identify all patients ≥18 years admitted with a primary diagnosis of AMI (weighted N = 13,030; unweighted N = 2,670) and concomitant OUD. Patients using other nonopiate illicit drugs were excluded. Propensity matching (1:1) yielded 2,253 well-matched pairs in which intergroup comparison of invasive revascularization strategies and cardiac outcomes were performed. The prevalence of OUD patients with AMI over the last decade has doubled, from 163 (2006) to 326 cases (2015) per 100,000 admissions for AMI. The OUD group underwent less cardiac catheterization (63.2% vs 72.2%; p <0.001), percutaneous coronary intervention (37.0% vs 48.5%; p <0.001) and drug-eluting stent placement (32.3% vs 19.5%; p <0.001) compared with non-OUD. No differences in in-hospital mortality/cardiogenic shock were noted. Among subgroup of ST-elevation myocardial infarction patients (26.2% of overall cohort), the OUD patients were less likely to receive percutaneous coronary intervention (67.9% vs 75.5%; p = 0.002), drug-eluting stent (31.4% vs 47.9%; p <0.001) with a significantly higher mortality (7.4% vs 4.3%), and cardiogenic shock (11.7% vs 7.9%). No differences in the frequency of coronary bypass grafting were noted in AMI or its subgroups. In conclusion, OUD patients presenting with AMI receive less invasive treatment compared with those without OUD. OUD patients presenting with ST-elevation myocardial infarction have worse in-hospital outcomes with increased mortality and cardiogenic shock.

Expert Consensus of Data Elements for Collection for Enhanced Recovery After Cardiac Surgery

BACKGROUND

Despite the emergence of Enhanced Recovery Protocols (ERPs) in cardiac surgery, there is no consensus on the essential elements for data reporting for quality improvement efforts, as well as accountability and standardization of outcome reporting across institutions. The aim of this study was to establish a consensus on essential data elements for cardiac ERAS®.

METHODS

A 2-round modified Delphi technique was utilized based on existing recommendations from the recently published ERAS® cardiac surgery consensus guidelines. Round 1 included a steering committee of 10 experts who oversaw formulation of a focused list of data elements into 3 main areas: Preoperative, intraoperative and postoperative. Round 2 consisted of a multidisciplinary, multinational, heterogenous group of 50 voting experts from across the United States and Europe. All participants evaluated their level of agreement with each data element using a 5-point Likert scale with consensus threshold of 70%.

RESULTS

In round 1, 17 data elements were considered essential (consensus >  = 70%, either positive or negative) and 6 were considered marginal (consensus <  = 70%, either positive or negative). In round 2, positive consensus was achieved for 15/17 (88.2%) data elements in the essential category, and all six data elements (100%) in the marginal category, indicating a high level of overall agreement.

CONCLUSION

This initial study, which identified 21 key data elements for collection in an ERAS® cardiac program, will aid clinicians in establishing a framework for evaluating the quality of their contemporary ERP processes and will allow acquisition of data to help benchmark performance metrics between hospitals.

Surgery for Endocarditis in Intravenous Drug Users

BACKGROUND

Few studies have analyzed outcomes after surgery for endocarditis in intravenous drug users (IVDUs). The aim of this study was to compare survival after surgery for endocarditis in IVDUs versus non-IVDUs. Secondary outcomes were the rates of reoperation, reinfection, and relapse to drug use.

METHODS

This population-based, observational cohort study included all patients who had undergone surgery for endocarditis at Karolinska University Hospital between 2002 and 2019. Patient data were collected from the institutional surgical database and medical charts. We used multivariable Cox regression to analyze associations between intravenous drug use and long-term survival.

RESULTS

Of the 510 study patients, 55 were IVDUs (11%) and 455 were not (89%). During a mean follow-up of 5.3 years (maximum, 17.1 years), 30 IVDUs (55%) and 133 non-IVDUs (29%) died. The 30-day mortality was 10.9% and 8.5%, respectively, for IVDUs and non-IVDUs (P = .53). Survival in IVDUs versus non-IVDUs at 1, 5, and 8 years was 76% versus 86%, 49% versus 76%, and 35% versus 68%, respectively (adjusted hazard ratio = 4.12; 95% confidence interval, 2.54-6.68; P < .001). The risk for reoperation was higher in IVDUs (adjusted hazard ratio = 3.47; 95% confidence interval, 1.74-6.89; P < .001). Forty-two IVDUs died or were reinfected (76%) and 49 died or returned to drug use (89%).

CONCLUSIONS

After surgery for endocarditis, IVDUs had substantially higher mortality and reoperation rates than did non-IVDUs. However, postoperative survival was comparable between groups, indicating that IVDUs manage surgery well. Prevention of relapse to drug use is of utmost importance in these patients.

Impact of opioid use disorders on outcomes and readmission following cardiac operations

OBJECTIVE

While opioid use disorder (OUD) has been previously associated with increased morbidity and resource use in cardiac operations, its impact on readmissions is understudied.

METHODS

Patients undergoing coronary artery bypass grafting and valve repair or replacement, excluding infective endocarditis, were identified in the 2010-16 Nationwide Readmissions Database. Using International Classification of Diseases 9/10, we tabulated OUD and other characteristics. Multivariable regression was used to adjust for differences.

RESULTS

Of an estimated 1 978 276 patients who had cardiac surgery, 5707 (0.3%) had OUD. During the study period, the prevalence of OUD increased threefold (0.15% in 2010 vs 0.53% in 2016, parametric trend<0.001). Patients with OUD were more likely to be younger (54.0 vs 66.0 years, p<0.001), insured by Medicaid (28.2 vs 6.2%, p<0.001) and of the lowest income quartile (33.6 vs 27.1%, p<0.001). After multivariable adjustment, OUD was associated with decreased mortality (1.5 vs 2.7%, p=0.001). Although these patients had similar rates of overall complications (36.1 vs 35.1%, p=0.363), they had increased thromboembolic (1.3 vs 0.8%, p<0.001) and infectious (4.1 vs 2.8%, p<0.001) events, as well as readmission at 30 days (19.0 vs 13.2%, p<0.001). While patients with OUD had similar hospitalisation costs ($50 766 vs $50 759, p=0.994), they did have longer hospitalisations (11.4 vs 10.3 days, p<0.001).

CONCLUSION

The prevalence of OUD among cardiac surgical patients has steeply increased over the past decade. Although the presence of OUD was not associated with excess mortality at index hospitalisation, it was predictive of 30-day readmission. Increased attention to predischarge interventions and care coordination may improve outcomes in this population.