Propensity-score analysis in thoracic surgery: When, why, and an introduction to how

Association between underweight status and chylothorax after esophagectomy for esophageal cancer: A propensity score-matched analysis

Objective

To use a nationwide database of hospitalizations to investigate underweight status as a risk factor for postesophagectomy complications.

Methods

We identified all patients who underwent esophagectomy with a diagnosis of esophageal cancer and known body mass index in the 2018-2020 Nationwide Readmissions Database. All hospital visits for esophagectomy and within 30 days of initial discharge were analyzed for postoperative complications, including chylothorax. Patients who were underweight were propensity score matched with patients who were not. Multivariable logistic regression was performed to identify complications that were significantly associated with underweight status.

Results

There were 1877 patients with esophageal cancer meeting inclusion criteria. Following propensity score matching, 433 patients who were underweight were matched to 433 patients who were not. In the multivariable model of the matched sample, which adjusted for age, sex, Charlson Comorbidity Index, history of chemotherapy or radiation therapy, and preoperative surgical feeding access, patients who were underweight were estimated to have 2.06 times the odds for chylothorax (95% confidence interval [CI], 1.07-4.25, P = .035). Underweight status was also significantly associated with acute bleed (odds ratio [OR], 1.52; 95% CI, 1.12-2.05, P = .007), pneumothorax (OR, 2.33; 95% CI, 1.19-4.85; P = .017), pneumonia (OR, 2.30; 95% CI, 1.53-3.50, P < .001), and in-hospital mortality (OR, 2.42; 95% CI, 1.31-4.69, P = .006).

Conclusions

Underweight status was found to be a risk factor for chylothorax after esophagectomy, which may have implications for perioperative care of esophageal cancer patients. Future studies should assess whether using feeding tubes or total parenteral nutrition preoperatively or thoracic duct ligation intraoperatively decreases risk of chylothorax among patients who were underweight.

Modifications to the video-assisted thoracoscopic surgery technique reduce 1-year mortality and postoperative complications in intrathoracic tumors

OBJECTIVE

The purpose of the study is to analyze the immediate outcomes and results of video-assisted thoracoscopic lobectomy and lung resection performed in the surgical department of the AOC between 2014 and 2018.

METHODS

For the period from 2014 to 2018, 118 patients with peripheral lung cancer were operated on in the surgical department of the AOC. The following operations were performed: lobectomy in 92 cases (78%), of which: upper lobectomy, 44 (47.8%); average lobectomy, 13 (14.1%); lower lobectomy, 32 (35%); bilobectomy, 3 (3.3%). All patients underwent extensive lymphadenectomy on the side of the operation. In 22 patients, for various reasons, preservation of thoracotomy was performed.

RESULTS

The absence of N0 lymph node damage was observed in 82 patients (70%), the first-order lymph node damage N1 in 13 (11%), N2 in 13 (11%), N3 in 5 (4%), and NX in 5 (4%). Histological examination revealed: squamous cell carcinoma - 35.1%, adenocarcinoma - 28.5%, undifferentiated carcinoma - 8.3%, NSCLC - 5.6%, NEO - 4.6%, sarcoma - 1.8%. At the same time, in 12.7% of patients, mts was detected - lung damage, and in 3.4%, malignant cells were not detected. Most patients were activated on the first day after surgery.

CONCLUSION

An analysis of the direct results of the study allows us to conclude that video-assisted thoracoscopic surgery is a highly effective, minimally invasive, safe method for treating peripheral lung cancer, which allows us to recommend it for wider use in oncological practice.

A retrospective study of laparoscopic, robotic-assisted, and open emergent/urgent cholecystectomy based on the PINC AI Healthcare Database 2017-2020

BACKGROUND

Robotic-assisted cholecystectomy (RAC) is becoming increasingly common, but the outcomes of emergent/urgent robotic-assisted cholecystectomies compared to emergent laparoscopic (LC) and open cholecystectomies (OC) remain understudied.

METHODS

The PINC AI Healthcare Database was queried to identify adults who underwent emergent or urgent (Em-Ur) cholecystectomy between January 1, 2017, and December 31, 2020. Immediate postoperative and 30-day outcomes were identified including intraoperative complications, transfusion, conversion, postoperative complication, and hospital length of stay. Propensity score matching was done to compare outcomes between Em-Ur robotic-assisted, laparoscopic, and open cholecystectomies Subgroup analyses were performed comparing RAC done with and without fluorescent imaging as well as comparing RAC and LC performed for patients with class 3 obesity (BMI ≥ 40 kg/m2).

RESULTS

RAC Em-Ur cholecystectomies are being performed with increasing frequency and is the most utilized modality for patients with class 3 obesity. There was no difference in intraoperative complications (0.3%), bile duct injury (0.2%), or postoperative outcomes between RAC and LC. LC had significantly shorter operating room times (96 min (75,128)) compared to RAC (120 min (90,150)). There was a significant lower rate of conversion to open in RAC (1.9%) relative to LC (3.2%) in both the overall population and the class 3 obesity sub-analysis (RAC-2.6% vs. LC-4.4%). There was no difference in outcomes in robotic-assisted cholecystectomies done with and without fluorescent imaging.

CONCLUSIONS

A comparison of propensity score-matched cohorts of emergent/urgent robotic-assisted and laparoscopic cholecystectomy indicates that robotic-assisted cholecystectomy is a safe alternative to laparoscopic cholecystectomy, and that both have superior outcomes to open cholecystectomies.

Association of survival with adjuvant radiotherapy for pN0 esophageal cancer

INTRODUCTION

This study was conducted to elucidate the link between adjuvant radiotherapy and survival in pathologic node-negative (pN0) esophageal cancer patients with upfront esophagectomy.

METHODS

From 2000 to 2016, patients with pN0 esophageal cancer who underwent upfront esophagectomy were selected from the Surveillance, Epidemiology, and End Results (SEER) database. The association of high-risk covariates with survival after adjuvant radiotherapy was evaluated using propensity score matching and multivariate analysis.

RESULTS

We identified 3197 patients, 321 (10.0%) underwent postoperative radiotherapy and 2876 (90.0%) underwent esophagectomy alone. In the unmatched cohort, postoperative radiotherapy was associated with a statistically significant but modest absolute decrease in survival outcomes (P < 0.001). In the matched cohort, the survival differences disappeared. Additionally, adjuvant radiotherapy was linked to a 5-year overall survival (OS) benefit for patients with the pT3-4N0 disease (34.8% vs. 27.7%; P = 0.008). Adjuvant radiotherapy for pT3-4N0 disease with tumor length ≥3 cm, adenocarcinoma, and evaluated lymph node count <12 was shown to independently function as a risk factor for improved OS, as per a multivariate analysis (P < 0.01).

CONCLUSIONS

This population-based trial showed that high-risk patients with pT3-4N0 esophageal cancer had better OS following upfront esophagectomy followed by radiotherapy therapy. This discovery may have major significance in the use of adjuvant radiotherapy following upfront esophagectomy in patients with pN0 esophageal cancer.

Temporal Trend and Clinical Outcomes in HIV and Non-HIV Patients following Liposuction: A Propensity-Matched Analysis

BACKGROUND

Because of the availability of highly active antiretroviral therapy, individuals infected with human immunodeficiency virus (HIV) are enjoying greater longevity with chronic conditions including abnormal adipose distribution. However, prior data on postoperative outcomes of liposuction in HIV-positive patients were limited by small sample size. Therefore, the authors aimed to compare differences in temporary trend, clinical characteristics, and outcomes between patients with and without HIV who underwent liposuction.

METHODS

The National Inpatient Sample database from 2010 to 2017 was queried to identify patients who underwent liposuction. Univariate, multivariate logistic regression and 1:4 propensity score-matched analyses were used to assess the primary outcomes (i.e., in-hospital mortality and postoperative outcomes) and secondary outcomes (i.e., discharge disposition, prolonged length of stay, and total cost).

RESULTS

Overall, 19,936 patients who underwent liposuction were identified, among whom 61 patients (0.31%) were infected with HIV. Patients with HIV were more likely to be male, insured by Medicare, and had more comorbidities and lower income. Unadjusted length of stay was longer among patients with HIV (OR, 1.81; 95% CI, 1.09 to 2.99; P = 0.020); nevertheless, multivariable models and propensity score-matched analysis demonstrated that patients with HIV were no more likely to have complications than the general population. This was also the case for length of stay and total costs.

CONCLUSIONS

The authors' findings indicated that patients with HIV who underwent liposuction did not experience an increased risk of major complication or mortality. Liposuction could be safely considered as a surgical treatment for HIV-positive patients with local fat deposition.

CLINICAL QUESTION/LEVEL OF EVIDENCE

Risk, II.

Impact of a digitized workflow for knee endoprothesis implantations on hospital-specific ratios

BACKGROUND: The continuous decrease of healthcare resources requires hospitals to improve efficiency while striving to improve quality standards that deliver better patient outcomes. OBJECTIVE: The objective of this study was to analyze whether the implementation of digital support systems during orthopedic surgery positively affected clinical processes and quality ratios. METHODS: A retrospective case-control study of 297 knee joint replacement procedures was conducted between 2015 and 2020. Thirty-five patients were allocated to the treatment and control groups after they were identified with exact matching and estimation of the propensity score. Both groups were balanced regarding the selected covariates. The effect of the surgical procedure manager (SPM) on the incidence of acute haemorrhagic anaemia between the two groups was evaluated with a t-test, and the odds ratio was calculated. RESULTS: SPM-supported surgery has no significant influence on the incidence of acute haemorrhagic anaemia but leads to significantly shorter hospital stay (1.93 days), changeover (4.14 minutes) and recovery room time (20.20 minutes). In addition, it reduces the standard deviation of operation room times. CONCLUSIONS: The study concludes that SPM enhances surgical efficiency and maintains quality outcomes. To overcome their increasing financial pressure hospital management should commercially evaluate the implementation of digital support systems.

Time-of-day dependent effects of midazolam administration on myocardial injury in non-cardiac surgery

Background

Animal studies have shown that midazolam can increase vulnerability to cardiac ischemia, potentially via circadian-mediated mechanisms. We hypothesized that perioperative midazolam administration is associated with an increased incidence of myocardial injury in patients undergoing non-cardiac surgery (MINS) and that circadian biology may underlie this relationship.

Methods

We analyzed intraoperative data from the Multicenter Perioperative Outcomes Group for the occurrence of MINS across 50 institutions from 2014 to 2019. The primary outcome was the occurrence of MINS. MINS was defined as having at least one troponin-I lab value ≥0.03 ng/ml from anesthesia start to 72 h after anesthesia end. To account for bias, propensity scores and inverse probability of treatment weighting were applied.

Results

A total of 1,773,118 cases were available for analysis. Of these subjects, 951,345 (53.7%) received midazolam perioperatively, and 16,404 (0.93%) met criteria for perioperative MINS. There was no association between perioperative midazolam administration and risk of MINS in the study population as a whole (odds ratio (OR) 0.98, confidence interval (CI) [0.94, 1.01]). However, we found a strong association between midazolam administration and risk of MINS when surgery occurred overnight (OR 3.52, CI [3.10, 4.00]) or when surgery occurred in ASA 1 or 2 patients (OR 1.25, CI [1.13, 1.39]).

Conclusion

Perioperative midazolam administration may not pose a significant risk for MINS occurrence. However, midazolam administration at night and in healthier patients could increase MINS, which warrants further clinical investigation with an emphasis on circadian biology.

Squamous cell carcinoma predicts worse prognosis than adenocarcinoma in stage IA lung cancer patients: A population-based propensity score matching analysis

Background

Although numerous studies have reported the association between histological types and the prognosis of IA non-small-cell lung cancer (NSCLC) patients, few studies have deeply investigated the impact of pathology on the outcome of NSCLC patients. In this study, we comprehensively explored whether the type of histology influenced the outcome of IA-stage NSCLC patients.

Methods

The study population was obtained from the Surveillance, Epidemiology, and End Results (SEER) program, which is supported by the National Cancer Institute of the United States. To avoid potential bias, the method of propensity score matching (PSM) was used to obtain a balanced cohort for further analysis.

Results

The results from univariate and multivariate regression models showed that lung squamous cell carcinoma (LSQCC) patients were at a significantly greater risk of undergoing shorter overall survival (OS) and lung cancer–specific survival (LCSS). After PSM analysis, LSQCC was still closely associated with a reduction in OS and LCSS. All of these suggested that the histological type was an independent prognostic factor for OS and LCSS.

Conclusion

Our study demonstrated that squamous cell carcinoma predicted worse OS and LCSS in IA-stage NSCLC patients compared with lung adenocarcinoma (LUAD). We suggest that the outcomes of LSQCC and LUAD are very different and that the two histological types should be differently analyzed.

Thoracoscopic versus open lobectomy in infants with congenital lung malformations: A multi-institutional propensity score analysis

PURPOSE

The impact of thoracoscopic surgery on outcomes in children with congenital lung malformations (CLM) remains controversial. The purpose of this study was to determine the effect of operative approach on perioperative outcomes in infants undergoing lobectomy for an asymptomatic CLM.

METHODS

After IRB approval, a retrospective cohort study was conducted on 506 children with a CLM resected at one of eleven children's hospitals over a seven-year period. Infants undergoing elective lobectomy were identified, and covariates were balanced based on operative approach using propensity scores with full matching. Outcomes were analyzed based on intention to treat with weighted conditional regression.

RESULTS

One hundred seventy-five infants met inclusion criteria. There were 67 (38.3%) open, 89 (50.9%) thoracoscopic, and 19 (10.9%) thoracoscopic-converted-to-open lobectomies. Thoracoscopic lobectomy was associated with significantly longer operative times (26 min, 95% CI 6-47 min, p = 0.012) but used less epidural anesthesia (OR 0.02, 95% CI 0.004-0.11, p<0.001) when compared to open lobectomy. There were no significant differences in intraoperative blood loss, postoperative complications, chest tube duration, or length of stay.

CONCLUSIONS

Thoracoscopy has become the most common operative approach for elective lobectomy in infants with asymptomatic CLMs. The non-inferiority of thoracoscopic lobectomy in postoperative outcomes supports its continued use as an alternative to open lobectomy.

LEVEL OF EVIDENCE

Treatment study, Level III.

Clinical and Economic Outcomes of Using Robotic Versus Hand-Held Staplers During Robotic Lobectomy

OBJECTIVE

During robotic lobectomy (RL), the surgeon can elect to use either robotic staplers or hand-held laparoscopic staplers. It is assumed that either will result in similar outcomes, while robotic staplers increase cost. We sought to compare perioperative outcomes and costs between RL cases that utilized robotic staplers versus hand-held staplers in real-world clinical practice.

METHODS

Patients who underwent an elective RL between October 2015 and December 2017 were identified in the Premier Hospital Perspective Database. Propensity score matching (PSM) analysis was performed to compare perioperative outcomes, healthcare resource utilization, and costs between cases using robotic staplers and hand-held staplers during RL.

RESULTS

In the PSM analysis, RL cases that fully utilized robotic staplers compared to hand-held staplers were associated with significantly lower risks of developing bleeding (5.6% vs 9.8%, P = 0.03) and conversion to open surgery (0.3% vs 5.9%, P = 0.004). Additionally, in a multivariable regression analysis, robotic stapler was associated with reduced risk for air leak (OR 0.70, 95% CI 0.50-0.98) and overall complications (OR 0.76, 95% CI 0.58-0.99). The total index hospitalization costs were comparable between the 2 groups (median [IQR], $21,667 [$16,860-$29,033] in robotic stapler vs $21,398 [$17,258-$29,406] in hand-held stapler, P = 0.22).

CONCLUSIONS

Among RL cases, utilization of robotic staplers was associated with significantly lower risks of perioperative bleeding, conversion, and possibly air leak and overall complications compared to RL cases utilizing hand-held staplers. The choice of stapler may have an impact on outcomes and robotic staplers do not increase total costs.

Propensity score matching in otolaryngologic literature: A systematic review and critical appraisal

BACKGROUND

Propensity score techniques can reduce confounding and bias in observational studies. Such analyses are able to measure and balance pre-determined covariates between treated and untreated groups, leading to results that can approximate those generated by randomized prospective studies when such trials are not feasible. The most commonly used propensity score -based analytic technique is propensity score matching (PSM). Although PSM popularity has continued to increase in medical literature, improper methodology or methodological reporting may lead to biased interpretation of treatment effects or limited scientific reproducibility and generalizability. In this study, we aim to characterize and assess the quality of PSM methodology reporting in high-impact otolaryngologic literature.

METHODS

PubMed and Embase based systematic review of the top 20 journals in otolaryngology, as measured by impact factor from the Journal Citations Reports from 2012 to 2018, for articles using PSM analysis throughout their publication history. Eligible articles were reviewed and assessed for quality and reporting of PSM methodology.

RESULTS

Our search yielded 101 studies, of which 92 were eligible for final analysis and review. The proportion of studies utilizing PSM increased significantly over time (p < 0.001). Nearly all studies (96.7%, n = 89) specified the covariates used to calculate propensity scores. Covariate balance was illustrated in 67.4% (n = 62) of studies, most frequently through p-values. A minority (17.4%, n = 16) of studies were found to be fully reproducible according to previously established criteria.

CONCLUSIONS

While PSM analysis is becoming increasingly prevalent in otolaryngologic literature, the quality of PSM methodology reporting can be improved. We provide potential recommendations for authors regarding optimal reporting for analyses using PSM.

Role of radiation therapy in node-negative esophageal cancer: A propensity-matched analysis

Background

This study investigated the prognostic impact of (neo‐)adjuvant radiation therapies in early stage esophageal cancer.

Methods

A retrospective analysis using the Surveillance, Epidemiology, and End Results (SEER) database was conducted from 2004 to 2016. Patients with pathologically staged T1‐4N0M0 esophageal cancer were divided into two treatment groups: (i) neoadjuvant radiotherapy followed by surgery; and (ii) upfront esophagectomy followed by adjuvant radiotherapy. Propensity scored match and Cox proportional hazards model were used to identify covariates associated with overall survival and cancer‐specific survival.

Results

There were 821 patients selected, of whom 588 (71.6%) received neoadjuvant radiotherapy and 233 (28.4%) received adjuvant radiotherapy. For the entire cohort, neoadjuvant radiotherapy was associated with a significantly benefit in five‐year survival outcomes compared with adjuvant radiotherapy (P < 0.01). After matching, the survival outcomes were still better for neoadjuvant radiotherapy than that of adjuvant treatment. Stratifying based on pathologic tumor status, neoadjuvant radiation was associated with improved CSS (five‐year survival 73.7% vs. 42.1%; P = 0.014) for localized (pT3‐4N0) disease. The Cox multivariate regression analysis revealed that the addition of neoadjuvant radiation for pT3‐4N0 diseases with tumor length ≥ 5 cm and squamous cell carcinoma, was a powerful prognostic factor for improved cancer‐specific survival (P < 0.01).

Conclusions

Compared with adjuvant radiotherapy, the addition of neoadjuvant radiation for pT3‐4N0 diseases has been associated with improved cancer‐specific survival in high‐risk patients. Studies on preoperative neoadjuvant therapies would be plausible in high‐risk esophageal cancer patients.

Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for clinical node-negative esophageal carcinoma

Background

The impact of neoadjuvant chemoradiotherapy (nCRT) on early stage esophageal cancer is unknown. Here, we compared the outcomes after esophagectomy alone or nCRT plus surgery for clinically staged node‐negative esophageal cancer.

Methods

We searched the Surveillance, Epidemiology, and End Results database for patients with clinically node‐negative (cN0) esophageal cancer from 2004 to 2016 who underwent surgery alone or nCRT plus surgery. Propensity score matching and Cox regression analysis were used to identify covariates associated with overall survival and cancer‐specific survival.

Results

A total of 1587 patients were retrospectively identified, of whom 49.8% (n = 791) received nCRT and 80.2% (n = 1273) were truly node‐negative diseases. For the entire cohort, surgery alone was associated with a statistically significant but modest absolute increase in survival outcomes (P < 0.01). After matching, nCRT was associated with improved five‐year overall survival for pT3‐4N0 (localized) disease (59.6% vs. 37.7%; P < 0.001) and pathological node‐positive disease (60.5% vs. 40.7%; P = 0.002). Cox multivariate regression analysis revealed that the addition of nCRT for truly node‐negative patients with tumor length ≥ 3 cm, pT1‐2N0 (early‐staged) and localized disease were independent risk factors for survival than surgery alone (P < 0.01).

Conclusions

Compared with surgery alone, patients with cN0 esophageal cancer with pathological node‐positive or localized true node‐negative disease gain a significant survival benefit from nCRT. However, nCRT plus surgery was associated with decreased survival for early‐staged true node‐negative patients. This finding may have significant implications on the use of neoadjuvant chemoradiation in patients with cN0 disease.

Comparison of the Short- and Long-term Outcomes of Video-assisted Thoracoscopic Surgery versus Open Thoracotomy Bronchial Sleeve Lobectomy for Central Lung Cancer: A Retrospective Propensity Score Matched Cohort Study

PURPOSE

The purpose of this study was to evaluate the short- and long-term outcomes of video-assisted thoracoscopic surgery (VATS) versus open thoracotomy bronchial sleeve lobectomy (BSL) for patients with central lung cancer.

METHODS

This is a retrospective cohort study. Perioperative outcomes and long-term survival of patients who underwent VATS versus open thoracotomy BSL for central lung cancer from June 2010 and June 2018 in the Western China Lung Cancer Database were compared using propensity score matching (PSM) between the two surgical approaches.

RESULTS

The retrospective study included 187 patients who divided into VATS group (n = 44) and open group (n = 143) according to surgical approach, and PSM resulted in 43 patients in each group, which were well matched by 11 potential prognostic factors. The VATS group was associated with lower overall incidence of postoperative complications (20.3% vs. 30.2%, P = 0.029), less postoperative drainage (875 ml [250-3960] vs. 1280 ml [100-4890], P = 0.039). The 5-year overall survival (OS) and disease-free survival (DFS) were comparable between the VATS and open groups (55.9% vs. 65.2% P = 0.836 and 54.1% vs. 60.2% P = 0.391, respectively) after matching. Multivariable adjusted analysis demonstrated that the surgical approach was not an independent favorable prognostic factor for OS (hazard ratio [HR] = 0.922; 95% confidence interval [CI], 0.427-1.993; P = 0.836) but just the pTNM stage (HR = 2.003; 95% CI 1.187-3.382; P = 0.009).

CONCLUSIONS

VATS BSL may achieve equivalent long-term outcomes for central lung cancer patients when comparing with open thoracotomy. Although slightly longer duration of surgery, VATS approach may be a feasible option for lung cancer patients requiring BSL.

Adjuvant radiotherapy for patients with pathologic node-negative esophageal carcinoma: A population based propensity matching analysis

BACKGROUND

The impact of adjuvant treatment for esophageal carcinoma with tumor-negative lymph nodes after upfront radical esophagectomy is still uncertain. This study investigated the effects of postoperative radiotherapy in pT1-3N0 esophageal carcinoma after radical resection.

METHOD

We retrospectively identified pT1-3N0M0 esophageal carcinoma patients between 2000 and 2016 from the Surveillance, Epidemiology, and End Results database. Patients with upfront esophagectomy were categorized as having received surgery alone (SA) and surgical resection followed by adjuvant radiotherapy (SA + RT). Propensity score matching, univariate and multivariate analysis were performed to compare overall survival (OS) and cause-specific survival (CSS).

RESULTS

A total of 2862 patients were identified, of whom 274 received SA + RT and 2588 received SA. The median follow-up was 60.4 months (95%CI, 58.7-62.1 months). The five-year OS and CSS were better for SA group compared with SA + RT group (P < 0.001, respectively). Furthermore, after matching, the OS and CSS were still significantly better for SA patients. For T subgroup analysis, postoperative radiotherapy was an independent prognostic factor only for pT1 patients with worse OS, without survival differences for pT2 and pT3 patients. However, after multivariate cox analysis, postoperative radiotherapy can provide significantly better OS for pT3 patients with tumor length ≥5 cm (P = 0.03; 95%CI, 0.29-0.94).

CONCLUSIONS

Among pT1-3N0M0 esophageal carcinoma patients, postoperative radiotherapy can provide significantly better OS for pT3 patients with tumor length ≥5 cm. However, there are no survival benefits for pT1-2 patients after SA + RT procedure. This finding may have significant implications on the use of adjuvant radiation in patients with pN0 disease.

Comparison of two cardiopulmonary bypass strategies with a miniaturized tubing system: a propensity score-based analysis

INTRODUCTION

The existing cardiopulmonary bypass tubing system has already been significantly improved in our hospital by reducing the priming volume; thus, we further employed a new cardiopulmonary bypass strategy in children based on a miniaturized cardiopulmonary bypass circuit. We aimed to compare the effectiveness of new and conventional strategies by analyzing the outcomes after congenital heart surgery.

METHODS

We performed a database analysis of all patients undergoing congenital heart surgery with cardiopulmonary bypass at Shenzhen Children's Hospital from 1 May 2015 to 30 June 2017. Propensity score matching was used to adjust for significant covariates, and multivariable regression models and stratified analysis were used to assess the association of cardiopulmonary bypass strategy with outcomes.

RESULTS

Of 925 total patients, 55.35% were in the conventional strategy group and 44.65% were in the new strategy group. After propensity score matching, there were 610 patients in total, with 305 patients in each group. In the multivariable regression models, the cardiopulmonary bypass strategy was not significantly associated with successful early extubation (p > 0.05), reintubation (p > 0.05), or nasal continuous positive airway pressure (p > 0.05) rates. The new strategy group had fewer hospital stays (p = 0.04) and intensive care unit stays (p < 0.05) compared with patients who underwent conventional strategy. The difference remained statistically significant (p < 0.05) when The Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery category was <3.

CONCLUSION

The implementation of a new cardiopulmonary bypass strategy, with selective ultrafiltration based on a miniaturized cardiopulmonary bypass circuit system, was safe and effective for children who underwent congenital heart surgery in a Chinese hospital. The new cardiopulmonary bypass strategy was associated with fewer hospital and intensive care unit stays.

Managing patients with acute type A aortic dissection and mesenteric malperfusion syndrome: A 20-year experience

OBJECTIVE

To assess outcomes of endovascular reperfusion followed by delayed open aortic repair for stable patients with acute type A aortic dissection and mesenteric malperfusion syndrome (mesMPS).

METHODS

Among 602 patients with acute type A aortic dissection who presented to our center from 1996 to 2017, all 82 (14%) with mesMPS underwent upfront endovascular fenestration/stenting. Primary outcomes were in-hospital mortality and long-term survival. Patients with acute type A aortic dissection with no malperfusion syndrome of any organ (n = 419) served as controls.

RESULTS

In-hospital mortality of all comers with mesMPS was 39%. After endovascular fenestration/stenting, 20 mesMPS patients (24%) died from organ failure and 11 patients (13%) died from aortic rupture before open aortic repair, 47 patients (58%) underwent aortic repair, and 4 patients (5%) survived without open repair. No patients died from aortic rupture during the second decade (2008-2017). The significant risk factors for death from organ failure after endovascular reperfusion were acute stroke (odds ratio, 23; 95% confidence interval, 4-144; P = .0008), gross bowel necrosis at laparotomy (odds ratio, 7; 95% confidence interval, 1.4-34; P = .016), and serum lactate ≥6 mmol/L (odds ratio, 13.5; 95% confidence interval, 2-97; P = .0097). There was no significant difference in operative mortality (2.1% vs 7.5%; P = .50) or long-term survival between patients with mesMPS who underwent open aortic repair after recovering from mesMPS and patients with no malperfusion syndrome.

CONCLUSIONS

In patients with acute type A aortic dissection with mesMPS, endovascular fenestration/stenting, and delayed open aortic repair achieved favorable short- and long-term outcomes. Surgeons should consider correcting mesenteric malperfusion before undertaking open aortic repair in patients with mesMPS, especially those with acute stroke, gross bowel necrosis at laparotomy, or serum lactate ≥6 mmol/L.

Is the Outcome of Pulmonary Resections due to Non-Small Cell Lung Cancer in Octogenarian Patients Worse?

Purpose: Lung cancer is one of the major sources of mortality in the elderly. This study was undertaken to assess the early and long-term results of surgical resection in patients older than 70 years of age by comparing the results of patients aged 70–79 years (group 1) with patients older than 80 years of age (group 2).

Methods: Data on patient age, gender, spirometry values, side, size, histology and stage of the tumor, surgical procedures, postoperative complications, Charlson comorbidity scores (CCS), and survival were collected.

Results: After 1–2 propensity score matching group 1 (70–79 years) included 84 and group 2 (age over 80) 42 cases. The multivariate analysis showed that CCS was the only significant factor affecting the development of complications (p = 0.003). The overall median and 5-year survival of all patients were 55 months and 42.5%, respectively. Although the survival of the elderly group 2 was higher than the first group, the difference did not reach significance (50 vs. 49 months, respectively).

Conclusion: The outcomes of surgery in terms of morbidity and mortality rates do not differ between the two age groups. The safety of pulmonary resections in the elderly group is comparable to patients under 70 years if the comorbidities are appropriately controlled. In addition, surgery provides satisfactory survival rates in both age groups.

Assessing the Quality of Microvascular Breast Reconstruction Performed in the Urban Safety-Net Setting: A Doubly Robust Regression Analysis

BACKGROUND

Safety-net hospitals serve vulnerable populations; however, care delivery may be of lower quality. Microvascular immediate breast reconstruction, relative to other breast reconstruction subtypes, is sensitive to the performance of safety-net hospitals and an important quality marker. The authors' aim was to assess the quality of care associated with safety-net hospital setting.

METHODS

The 2012 to 2014 National Inpatient Sample was used to identify patients who underwent microvascular immediate breast reconstruction after mastectomy. Primary outcomes of interest were rates of medical complications, surgical inpatient complications, and prolonged length of stay. A doubly-robust approach (i.e., propensity score and multivariate regression) was used to analyze the impact of patient and hospital-level characteristics on outcomes.

RESULTS

A total of 858 patients constituted our analytic cohort following propensity matching. There were no significant differences in the odds of surgical and medical inpatient complications among safety-net hospital patients relative to their matched counterparts. Black (OR, 2.95; p < 0.001) and uninsured patients (OR, 2.623; p = 0.032) had higher odds of surgical inpatient complications. Safety-net hospitals (OR, 1.745; p = 0.005), large bedsize hospitals (OR, 2.170; p = 0.023), and Medicaid patients (OR, 1.973; p = 0.008) had higher odds of prolonged length of stay.

CONCLUSIONS

Safety-net hospitals had comparable odds of adverse clinical outcomes but higher odds of prolonged length of stay, relative to non-safety-net hospitals. Institution-level deficiencies in staffing and clinical processes of care might underpin the latter. Ongoing financial support of these institutions will ensure delivery of needed breast cancer care to economically disadvantaged patients.

CLINICAL QUESTION/LEVEL OF EVIDENCE

Therapeutic, III.

Evaluation of the real-world impact of rotational thromboelastometry-guided transfusion protocol in patients undergoing proximal aortic surgery

BACKGROUND

Complex aortic procedures are potentially associated with important blood loss and coagulopathy. The aim of this study was to assess the impact of rotational thromboelastometry (ROTEM, Tem International GmBH, Munich, Germany) on transfusion requirements after proximal aortic operations in a real-world setting.

METHODS

This single-center retrospective analysis based on 385 consecutive patients undergoing cardiac surgeries involving the aortic root, ascending aorta, or aortic arch compared 197 controls managed according to routine transfusion protocol before the introduction of the ROTEM in 2012 with 188 patients operated afterward. With the use of a 1:1 propensity score match, 224 patients were included in paired analysis (112 in each group). The primary end point was erythrocytes transfusion rate. The secondary end points comprised the transfusion of other allogeneic blood products, number of units transfused, postoperative blood loss, massive transfusion rate, and use of other hemostatic products.

RESULTS

ROTEM implementation was associated with a trend toward reduction in the rate of erythrocytes transfusion (57% vs 46%, P = .08) and a decreased median number of units transfused for erythrocytes (1.0 [0.0-4.0] unit vs 0.0 [0.0-2.0] unit, P = .03) and plasma (0.0 [0.0-4.0] unit vs 0.0 [0.0-2.0] unit, P = .04). After sensitivity analysis, ROTEM displayed a comparable rate of erythrocytes transfusion (58% vs 47%, P = .15).

CONCLUSIONS

In a real-world setting, ROTEM-based algorithm implementation could help reduce excess erythrocytes transfusion for complex aortic procedures. We advocate for a strict adherence and concerted team effort to maximize the benefits of such addition to patients' management.

Recombinant Factor VIIa Is Associated With Increased Thrombotic Complications in Pediatric Cardiac Surgery Patients

BACKGROUND

Recombinant factor VIIa (rFVIIa) is routinely used as an off-label hemostatic agent in children undergoing cardiac surgery. Despite evidence that rFVIIa use is associated with an increased incidence of thrombotic complications in adult cardiac surgery, the safety of rFVIIa as a rescue hemostatic agent in the pediatric cardiac surgical population is less definitively delineated. In this retrospective study, we used propensity score matching to compare the incidence of thrombotic complications between children treated with rFVIIa and their matched controls.

METHODS

We retrospectively reviewed medical records and pharmacy data from all neonates and children who underwent congenital cardiac surgery between May 1, 2011, and October 31, 2013, at Boston Children's Hospital, and identified those who received rFVIIa during the perioperative period. Using existing knowledge, we chose 10 factors associated with bleeding after cardiac surgery to be used in our propensity score: age, sex, body weight, neonates, prematurity, previous sternotomy, cardiopulmonary bypass time, deep hypothermic circulatory arrest time, aortic cross-clamp time, and the operative surgeon. We then used propensity-matched analysis to match children treated with rFVIIa with 2 controls. The primary outcome was thrombotic complications. Secondary outcomes included reexploration for bleeding, length of cardiac intensive care unit stay, length of hospital stay, and 30-day mortality.

RESULTS

One hundred forty-nine patients received perioperative rFVIIa during the study period. Propensity matching yielded 143 rFVIIa patients matched to 2 control patients each (n = 286). Three control patients were found to have received rFVIIa during the perioperative course and were removed from the analysis, for a total of 283 control patients. The administration of rFVIIa was associated with an increased incidence of thrombotic complications (20% vs 8%; odds ratio [OR]: 3.9 [95% confidence interval {CI}: 2.6-5.9], P < .001). Administration of rFVIIa was associated with a prolonged median length of cardiac intensive care unit stay (8 days [interquartile range {IQR}: 4-24] vs 5 days [IQR: 2-10], P < .001) and prolonged length of hospital stay (20 [IQR: 9-44] vs 11 days [IQR: 7-23], P < .001). No difference in reexploration for bleeding (rFVII = 14% vs controls = 9%; OR: 1.7 [95% CI, 0.92-3.1], P = .12) or 30-day mortality was observed (8% vs 6%; OR 1.3 [95% CI, 0.60-2.89], P = .51).

CONCLUSIONS

This retrospective analysis confirmed that perioperative administration of rFVIIa is associated with an increased incidence of postoperative thrombotic complications in neonates and children undergoing cardiac surgery, without increase in 30-day mortality. In conclusion, rFVIIa should be used with extreme caution in pediatric patients undergoing cardiac surgery.

Surgery versus percutaneous transcatheter embolization for pulmonary arteriovenous malformation: Analysis of a national inpatient database in Japan

OBJECTIVE

Although the treatment options for pulmonary arteriovenous malformation are surgery and percutaneous transcatheter embolization, no study has compared the outcomes between these 2 treatments.

METHODS

From the Japanese Diagnosis Procedure Combination database, the medical records of 996 patients who received treatment for pulmonary arteriovenous malformation between 2010 and 2015 were retrospectively reviewed. We created balanced groups for surgery or percutaneous transcatheter embolization using propensity scoring. The primary outcome was the rate of reintervention for pulmonary arteriovenous malformation, and the secondary outcomes were composite complications and postoperative length of stay. Patients who had any 1 of the complications during hospitalization were considered to have experienced a composite complication.

RESULTS

Of the total sample, 211 patients underwent surgery and 785 patients underwent percutaneous transcatheter embolization. By using 1-to-1 propensity score matching, 202 pairs were selected. Compared with percutaneous transcatheter embolization, surgery was associated with a significantly higher proportion of composite complications (6.9% vs 2.0%, P = .027) and longer postoperative length of hospital stay (median, 6 vs 2 days, P < .01). However, surgery resulted in a significantly lower rate of reintervention for pulmonary arteriovenous malformation (2.1% vs 8.3% at 2 years; P < .01).

CONCLUSIONS

Percutaneous transcatheter embolization had the advantage in composite complications and shorter postoperative length of stay compared with surgery, but surgery had higher curability than percutaneous transcatheter embolization. Surgery may be considered as a therapeutic option for patients with lesions that can be completely resected and are difficult to treat with percutaneous transcatheter embolization.

Tips and tricks of the propensity score methods in the thoracic surgery research

Without randomization, the differences in the distribution of baseline covariates can confound the evaluation of outcomes between the groups. To estimate the causal effects of treatment on the results randomized clinical trials (RCT) are routinely used. When RCT are not feasible for cost, time, and ethical issues, the effect of treatment on an inevitable outcome could be investigated by using a non-experimental study design. The propensity score (PS) refers to the individual probability, for a subject involved in a study, of receiving a new treatment rather than the control treatment. PS matching is a valuable and increasingly popular tool for dealing with observational data and non-random treatment assignment. Nevertheless, surgeons need to proceed with caution and apply PS methods appropriately.