Elsevier

American Journal of Otolaryngology

Volume 39, Issue 2, March–April 2018, Pages 192-196
American Journal of Otolaryngology

Original contribution
ACS NSQIP risk calculator reliability in head and neck oncology: The effect of prior chemoradiation on NSQIP risk estimates following laryngectomy

https://doi.org/10.1016/j.amjoto.2017.11.005Get rights and content

Abstract

Purpose

To determine whether inclusion of chemoradiation history increases estimated risk for complications following total laryngectomy using the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) Surgical Risk Calculator.

Materials and methods

A retrospective review of 96 patients with laryngeal cancer, approximately half of who had received prior chemoradiation, who underwent laryngectomy between January 2010 and December 2014. NSQIP estimates were calculated and compared to actual event occurrence using receiver operating characteristic (ROC) curves, Brier scores, and risk estimates.

Results

Patients who had received prior chemoradiation were at significantly greater risk for complication postoperatively (OR = 2.63, 95% CI = 1.145–6.043). NSQIP Calculator discriminability and accuracy were generally poor for this sample. While NSQIP estimates significantly predicted risk for any postoperative complication, pneumonia, and discharge to nursing care for primary laryngectomy patients, predictive capability was lost among salvage laryngectomy patients. NSQIP adjustments to both Somewhat Higher and Significantly Higher Risk categories did not improve predictive capability. Of the risk factors considered by NSQIP, preoperative functional status (p = 0.041), age at time of surgery (p < 0.008), and inclusion of neck dissection (p = 0.035) emerged as significant predictors of actual postoperative complications, though again estimates lost significance among salvage laryngectomy patients.

Conclusions

The NSQIP Calculator may be poorly calibrated to estimate postoperative complication risk for patients previously exposed to chemoradiation undergoing salvage laryngectomy. Caution should be used when estimating postoperative risk among patients undergoing salvage procedures, especially those of older age, poorer functional status, and those requiring neck dissection.

Introduction

Among patients with head and neck cancer, estimates of postoperative wound complications following total laryngectomy vary, ranging from 7 to 41% [1]. These complications include infections, dehiscence, and pharyngocutaneous fistulas, which can increase patient morbidity and mortality rates, leading to extended Intensive Care Unit stays, longer hospital admissions and increased healthcare costs [2]. Many factors have previously been associated with postoperative complications, including previous radiotherapy, preoperative tracheostomy, radical neck dissection, and extensive surgery and flap reconstruction [2]. The majority of laryngectomies are reserved for patients who fail organ sparing therapy with chemoradiation [3]. As such, complication rates are higher compared to those seen among patients undergoing primary laryngectomy [4], [5]. Prior radiation is considered an independent predictor for both local and fistula complications [2]. Although radiation has been established as a risk factor for complications after total laryngectomy, it has not typically been used in risk calculators to predict post-operative complications [6], [7].

While the concept of risk adjustment is not new, the use of standardized and database-derived calculators to perform risk adjustment has become more widespread. For patients and surgeons, these estimates allow for better informed consent and understanding of the risks of surgery. They are also used to drive compensation as systems like the Centers for Medicare and Medicaid Services (CMS) incentivize surgeons to discuss empirically derived risks with patients prior to performing operative procedures [8]. Perioperative risk calculation is becoming increasingly popular, especially with the continued development of easily accessible methods that may be completed in the presence of the patient. One readily available risk calculator is the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) Surgical Risk Calculator. Using standardized clinical data of preoperative risk factors and postoperative complications collected from over 2.7 million operations from > 500 hospitals in the US, the NSQIP Surgical Risk Calculator estimates the chance of various unfavorable outcomes (e.g., infection, cardiac complication, death, etc.) within 30-days following surgery (Table 1) [8], [9]. These estimates are based on outcomes that are continually refined as a greater diversity of patients and procedures are entered into the database. Individual risk is estimated based upon patient-provided information about prior health history, compared to a database of information from patients who had a similar surgical procedure, and a determination about the likelihood of outcomes and complications is made [8]. If the surgeon believes that reported risks are underestimated, the “Surgeon Adjustment of Risk” option can be used to adjust predicted risks either to the first or second standard deviation above the mean for a particular CPT code. Currently, the NSQIP Surgical Risk Calculator does not consider prior radiotherapy into its risk estimation for total laryngectomy. Therefore, enhancing our ability to predict potential postoperative complications is paramount.

We sought to determine if receipt of prior radiotherapy with or without chemotherapy would impact risk estimates derived from the NSQIP Surgical Risk Calculator for head and neck cancer patients undergoing salvage laryngectomy. We compared NSQIP-predicted rates of complications with actual complication rates among a sample of head and neck cancer patients who had undergone laryngectomy. We examined radiation-naïve patients undergoing laryngectomy as a comparison group.

Section snippets

Patient sample

After Institutional Review Board approval was obtained with complete waiver of consent, a retrospective review of cases completed January 2010 through December 2014 was conducted at an academic and private institution. Patients were included if they had a confirmed diagnosis of cancer of the head and neck region and had undergone a partial or total laryngectomy. CPT codes searched included those for total and subtotal laryngectomy with and without neck dissection (CPT 31360, 31365, 31367,

Sample characteristics

Perioperative and postoperative characteristics of the sample are provided in Table 2. Patients were either of independent or partially dependent functional status. None had disseminated cancer, acute renal failure, or systemic sepsis within 48 h prior to procedure. In the primary laryngectomy group, one patient was ventilator dependent. In the salvage laryngectomy group, one patient required dialysis, one had ascites, one required steroid for a chronic condition, and one had experienced

Discussion

Here we demonstrate in a sample of 96 patients that those undergoing salvage laryngectomy (n = 48) were significantly more likely to experience any complication postoperatively compared to those undergoing primary laryngectomy. While the NSQIP demonstrated good predictive capability and calibration in the primary laryngectomy group under study, NSQIP Calculator estimates lost predictive capability in the salvage surgery cohort when estimating risk of any complication occurring postoperatively.

Conclusions

These data provide the initial identification of preoperative risk factors with the greatest potential to negatively impact surgical outcomes following laryngectomy. Caution should be used when estimating postoperative risk among head and neck patients undergoing salvage laryngectomy, especially those of older age, poorer functional status and those requiring neck dissections. Given the evidence linking prior chemoradiation to serious postoperative complications for head and neck cancer

Acknowledgements

The authors thank Erik Rasmussen MD for his help in collecting and managing the data for the study. The authors also thank the health information management staff at their facilities for providing assistance with locating records used in analyses. This research did not receive any specific funding from agencies in the public, commercial, or not-for-profit sectors.

References (18)

  • S.R. Schwartz et al.

    Predictors of wound complications after laryngectomy: a study of over 2000 patients

    Otolaryngol Head Neck Surg

    (2004)
  • I. Ganly et al.

    Analysis of postoperative complications of open partial laryngectomy

    Head Neck

    (2009)
  • E.E. Vokes

    Competing roads to larynx preservation

    J Clin Oncol

    (2013)
  • I.M. Agra et al.

    Postoperative complications after en bloc salvage surgery for head and neck cancer

    Arch Otolaryngol Head Neck Surg

    (2003)
  • M. Stankovic et al.

    Influential factors, complications and survival rate of primary and salvage total laryngectomy for advanced laryngeal cancer

    Coll Antropol

    (2012)
  • G.A. Croll et al.

    Primary radiotherapy with surgery in reserve for advanced laryngeal carcinoma. Results and complications

    Eur J Surg Oncol

    (1989)
  • A.W. McCombe et al.

    Radiotherapy and complications of laryngectomy

    J Laryngol Otol

    (1993)
  • K.Y. Bilimoria et al.

    Development and evaluation of the universal ACS NSQIP surgical risk calculator: a decision aid and informed consent tool for patients and surgeons

    J Am Coll Surg

    (2013)
  • American College of Surgeons

    ACS NSQIP surgical risk calculator

There are more references available in the full text version of this article.

Cited by (0)

Disclosures: None of the authors have any disclosures, financial or otherwise, to make as concerns the content matter of this study.

View full text