Correlation of hemoglobin A1C level with surgical outcomes: Can tight perioperative glucose control reduce infection and cardiac events?

Investigating glycemic control in patients undergoing lower extremity bypass within an enhanced recovery pathway at a single institution

BACKGROUND

Enhanced recovery pathways (ERPs) aim to lower perioperative stress to facilitate recovery. Limited fasting combined with carbohydrate loading is a common ERP element. The effect of limited fasting has not been elucidated in patients with diabetes. Given the known deleterious effects of poor glycemic control in the perioperative period, such as increased rates of surgical site infection, the associations of preoperative limited fasting with perioperative glycemic control and early outcomes after lower extremity bypass (LEB) were investigated.

METHODS

A single institutional retrospective review of patients who underwent infrainguinal LEB from 2016 to 2022 was performed. The ERP was initiated in May 2018. Patients were stratified by diabetes diagnosis and preoperative hemoglobin A1C (HbA1C) levels. Perioperative glycemic control was compared between the limited fasting and traditional fasting patients (nil per os at midnight). Limited fasting was defined as a clear liquid diet until 2 hours before surgery with recommended carbohydrate loading consisting of 400 cc of a clear sports drink (approximately 30 g of carbohydrates). All limited fasting patients were within the ERP. Early perioperative hyperglycemia (EPH) was defined as blood glucose of >180 mg/dL within the first 24 hours of surgery. Perioperative outcomes such as surgical site infection, readmission, reinterventions, and complications were also compared.

RESULTS

A total of 393 patients were included (limited fasting patients N = 135; traditional fasting patients N = 258). A trend toward EPH was seen in all limited fasting groups. Evaluating limited fasting within diabetic patients revealed that 74.5% of limited fasting-diabetic patients had EPH compared with 49.6% of traditional fasting-diabetic patients (P = .001). When stratified by the HbA1C level, a significantly higher rate of EPH was seen in the HbA1c >8.0% groups, with 90.5% in the limited fasting patients compared with 67.9% in traditional fasting patients (P = .05). Limited fasting-diabetic patients experience a longer postoperative length of stay at 5.0 days (interquartile range: 3, 9) vs 4.0 days (2, 6) in nondiabetic patients (P = .016).

CONCLUSIONS

ERP limited fasting was associated with early perioperative hyperglycemia after LEB, particularly in patients with HbA1C >8.0%. Due to the high prevalence of diabetic patients undergoing LEB under ERP, the role of limited fasting and common glycemic elements of ERP may need to be re-evaluated in this subpopulation.

The impact of postoperative blood glucose levels on complications and prognosis after esophagectomy in patients with esophageal cancer

PURPOSE

This study aimed to characterize postoperative blood glucose fluctuation in patients who underwent esophagectomy for esophageal cancer, and to define its impact on complications and prognosis.

METHODS

The subjects of this retrospective study were 284 patients who underwent esophagectomy at Osaka University Hospital between 2015 and 2017. Data analyzed included clinicopathological background, the immediate postoperative blood glucose level (IPBG), postoperative blood glucose variability (PBGV), insulin dosage, postoperative complications, and prognosis.

RESULTS

The median IPBG and PBGV were 170 (64-260) mg/dl and 64.5 (11-217) mg/dl, respectively. Postoperative pneumonia was more common in patients with PBGV > 100 mg/dl (P = 0.015). Patients with IPBG < 170 mg/dl had significantly worse 5-year overall survival (OS) and 5-year recurrence-free survival (RFS) than those with IPBG > 170 mg/dl (54.5% vs. 80.4%, respectively, [P < 0.001] and 44.3% vs. 69.3%, respectively, [P = 0.001]). The 5-year OS rates were 43.5%, 68.3%, 80.6%, and 79.0% for patients with IPBG < 154, 154-170, 170-190, and ≥ 190 mg/dl, respectively. The corresponding 5-year RFS rates were 38.1%, 52.4%, 77.0%, and 61.3%, respectively. Multivariate analysis revealed that IPBG < 154 mg/dl and pathological stage were independent poor prognostic factors for OS.

CONCLUSION

PBGV was associated with postoperative pneumonia, and low IPBG was an independent poor prognostic factor for patients with esophageal cancer.

The impact of weight change on intra-abdominal and hernia volumes

BACKGROUND

Weight loss is often encouraged or required before open ventral hernia repair. This study evaluates the impact of weight change on total, intra-abdominal, subcutaneous, and hernia volume.

METHODS

Patients who underwent open ventral hernia repair from 2007 to 2018 with two preoperative computed tomography scans were identified. Scans were reviewed using 3D volumetric software. Demographics, operative characteristics, and outcomes were evaluated. The impact of weight change on intra-abdominal, subcutaneous, and hernia volume was assessed using Spearman's correlation coefficients and linear regression models.

RESULTS

A total of 250 patients met the criteria with a mean defect area of 155.6 ± 155.4 cm², subcutaneous volume of 6,800.0 ± 3,868.8 cm³, hernia volume of 915.7 ± 1,234.5 cm³, intra-abdominal volume equaling 4,250.2 ± 2,118.1 cm³, and time between computed tomography scans 13.9 ± 11.0 months. Weight change was associated with change in hernia, intra-abdominal, total, and subcutaneous volume (Spearman's correlation coefficients 0.17, 0.48, 0.51, 0.45, respectively, P ≤ 0.03 all values) and not associated in hernia length, width, or area (P ≥ 0.18 all values). A Δ5 kg was significantly associated with Δintra-abdominal volume (164.1 ± 30.0 cm³/Δ5 kg,P < .0001), Δtotal volume (209.9 ± 33.0 cm³/Δ5 kg, P < .0001), and Δsubcutaneous volume (234.4 ± 50.8 cm³/Δ5 kg, P < .0001). Per Δ5 kg, male patients had more than double the Δintra-abdominal, Δtotal, and Δsubcutaneous volume than did female patients. A weight change of 5 kg to10 kg was associated with approximately double the change in computed tomography parameters/Δ5 kg than any weight change after 10 kg. Regardless of weight change, all measured hernia parameters increased over time, with mean hernia volume of +40.6 ± 94.9 cm³/mo and area of +7.8 ± 13.3 cm²/mo (Spearman's correlation coefficient -0.03 to 0.07, P value 0.37-0.96).

CONCLUSION

Weight change is linearly correlated with intra-abdominal and subcutaneous fat gain or loss. Males show greater abdominal-related response to weight gain or loss. Hernia dimensions increase over time regardless of weight change.

Preoperative Glycosylated Hemoglobin Levels Predict Anastomotic Leak After Esophagectomy with Cervical Esophagogastric Anastomosis

BACKGROUND

Patients with diabetes are considered at increased risk of delayed wound healing and infectious complications, yet the relationship between diabetes and anastomotic leak (AL) remains unclear. Given that glycosylated hemoglobin (HbA1c) is a validated indicator of the long-term glycemic state, we evaluated the relationship between preoperative HbA1c levels and AL after esophagectomy.

METHODS

We assessed 300 consecutive patients who underwent esophagectomy reconstructed with cervical esophagogastric anastomosis between 2011 and 2015. HbA1c levels were measured within 90 days before esophagectomy. We performed comparison between the patients with and without diabetes. In addition, the predictive factors for AL, as well as the relationship between HbA1c levels and AL, were investigated.

RESULTS

Among the 300 patients, 35 had diabetes. The overall prevalence of AL was 11.7%, and patients with diabetes had a higher prevalence of AL than those without (p = 0.045). In univariate analysis, we identified diabetes, HbA1c level, and hand-sewn anastomosis as risk factors for AL significantly (p = 0.033, 0.009, and 0.011, respectively), but we also found previous smoking history, chronic hepatic disease, and supracarinal tumor location also showed tendencies to be risk factors (p = 0.057, 0.055, and 0.064, respectively). Multivariate logistic regression analysis indicated that chronic hepatic disease (p = 0.048), increased HbA1c level (p = 0.011), and hand-sewn anastomosis (p = 0.021) were independent risk factors for AL.

CONCLUSIONS

Preoperative HbA1c level was significantly associated with the development of AL after cervical esophagogastric anastomosis. We recommend preoperative HbA1c screening for all patients scheduled to undergo esophagectomy.

Prospective Study of Postoperative Glycemic Control with a Standardized Insulin Infusion Protocol after Infrainguinal Bypass and Open Abdominal Aortic Aneurysm Repair

BACKGROUND

The aim of this study is to examine the effect of moderate postoperative glycemic control in diabetic and nondiabetic patients undergoing infrainguinal bypass (INFRA) or open abdominal aortic aneurysm (OAAA) repair.

METHODS

In a single center prospective study, we investigated postoperative glycemic control using a standardized insulin infusion protocol after elective INFRA bypass (n = 53, 62%) and OAAA repair (n = 33, 38%) between January 2013 and March 2015. The primary end point was optimal glycemic control, defined as having ≥85% of blood glucose values within the 80-150 mg/dL target range. Suboptimal glycemic control was defined as <85% of blood glucose values within the blood glucose target range. Secondary end points included in-hospital and 30-day surgical site infection (SSI) rates, composite adverse events, length of stay (LOS), and hospital cost.

RESULTS

Optimal glycemic control was achieved more commonly after OAAA repair than INFRA bypass (85% vs. 64%, P = 0.04). Moderate hypoglycemia (<70 mg/dL) was observed in 32 (37%) patients, while severe hypoglycemia (<50 mg/dL) was observed in 6 (7%) patients. SSI at 30 days was more common after INFRA bypass (n = 15, 29%) than OAAA repair (n = 2, 6%) (P = 0.01). In-hospital (6% vs. 6%, P = 1.0) and 30-day (24% vs. 22%, P = 1.0) SSI rates were similar for optimal versus suboptimal glycemic control patients after INFRA bypass. In-hospital (4% vs. 0%, P = 1.0) and 30-day (4% vs. 0%, P = 1.0) SSI rates were similar for optimal versus suboptimal glycemic control patients after OAAA repair. The percentage of blood glucose > 250 mg/dL was similar for patients with and without SSI (3% vs. 2%, P = 0.36). Adverse cardiac and pulmonary events after INFRA bypass were similar between groups (9% vs. 21%, P = 0.23; 0% vs. 5%, P = 0.36, respectively). Adverse cardiac and pulmonary events after OAAA repair were similar between groups (2% vs. 0%, P = 1.0; 4% vs. 0%, P = 1.0, respectively). Mean LOS was significantly lower in patients with optimal glycemic control after INFRA bypass (4.2 vs. 7.3 days, P = 0.02). Mean LOS was similar after OAAA repair for patients with optimal and suboptimal control (5.8 vs. 6.4 days, P = 0.46). Inpatient hospital costs after INFRA bypass were lower for the group with optimal (median $25,012, interquartile range [IQ] range $21,726-28,331) versus suboptimal glycemic control (median $28,944, IQ range 24,773-41,270, P = 0.02).

CONCLUSIONS

Postoperative hyperglycemia is common after INFRA bypass and OAAA repair and can be effectively ameliorated with an insulin infusion protocol. The protocol was low risk with reduced LOS and cost after INFRA bypass. Complications including SSI were not reduced in patients with optimal perioperative glycemic control.

Postoperative complications of total laryngectomy in diabetic patients

OBJECTIVES/HYPOTHESIS

Analyze postoperative complications of total laryngectomies (TL) in patients with diabetes mellitus and apply these data toward preoperative management of diabetic patients undergoing TL.

STUDY DESIGN

Retrospective analysis of surgical outcomes database METHODS: The American College of Surgeons National Surgical Quality Improvement Program database was queried for TL performed between 2005 and 2013. Comparison of demographics, preoperative comorbidities, and postoperative outcomes in a diabetic versus a nondiabetic cohort was conducted using SPSS software. Equal distribution of demographics and preoperative comorbidities was ensured between the diabetic and nondiabetic cohorts using a propensity-matching algorithm.

RESULTS

Following propensity matching there were 495 TL patients included in this study. Among these, 110 (22.2%) were diabetic and 385 (77.8%) were nondiabetic. The only comorbidities associated with the diabetic cohort after matching were obesity, previous percutaneous coronary intervention, and hypertension. The rates of superficial surgical site infections were higher in the diabetic patient cohort (10.9% vs. 4.7%, P = .022). There were no significant differences in unplanned readmission (12.7% vs. 9.1%, P = .260), reoperation (14.3% vs. 15.1%, P = .864), and mortality (1.8% vs. 2.6%, P = 1.000) between the diabetic and the nondiabetic groups. Multivariate regression utilizing preoperative variables not accounted for by propensity matching showed that superficial surgical site infections were still higher in the diabetes cohort (odds ratio: 3.371, P = .021).

CONCLUSIONS

Diabetic patients undergoing TL showed an increased incidence of superficial surgical site infections postoperatively. There were no other significant differences in readmission, reoperation, and mortality.

LEVEL OF EVIDENCE

4 Laryngoscope, 127:2247-2251, 2017.

Effect of diabetes mellitus on postoperative endoscopic sinus surgery outcomes

BACKGROUND

Endoscopic sinus surgery (ESS) has become the treatment of choice for a variety of nasal conditions. The purpose of this study was to analyze the effect of diabetes mellitus (DM) on postoperative outcomes in ESS.

METHODS

Data on endoscopic sinus surgery performed from 2005 to 2013 were collected from the American College of Surgeons National Surgical Quality Improvement (ACS-NSQIP) database. Two groups were created, based on the presence of a DM diagnosis, and were analyzed for preoperative variables, comorbidities, and postoperative complications using SPSS statistical software.

RESULTS

There were 644 patients included in the analysis, 85 of whom (13.2%) had a diagnosis of DM. Patients with DM were more likely to have higher rates of preoperative dyspnea and hypertension. After accounting for confounding factors, DM patients undergoing ESS were at higher risk of overall medical complications, pneumonia, unplanned reintubation, ventilator use of >48 hours, and mortality. However, after separating patients into outpatient and inpatient groups, DM was found to be an independent predictor of urinary tract infection in outpatients and of ventilator use >48 hours in inpatients.

CONCLUSION

DM patients undergoing ESS are at increased risk for postoperative medical complications. However, DM does not appear to increase the postoperative surgical complication rate in this population. Furthermore, DM does not appear to have an impact on ESS mortality, readmission, or reoperation rates.