Now Reading
Preoperative Serum Albumin Predicts Wound Dehiscence but Not Infection After Surgery for Oral Squamous Cell Carcinoma

Preoperative Serum Albumin Predicts Wound Dehiscence but Not Infection After Surgery for Oral Squamous Cell Carcinoma

Preoperative Serum Albumin Predicts Wound Dehiscence but Not Infection After Surgery for Oral Squamous Cell Carcinoma

Journal of Oral and Maxillofacial Surgery, 2021-06-01, Volume 79, Issue 6, Pages 1355-1363, Copyright © 2020 American Association of Oral and Maxillofacial Surgeons


Inadequate nutrition is common in individuals diagnosed with cancer. The present study evaluated the association between preoperative albumin and postoperative complications in otherwise healthy patients presenting with newly diagnosed squamous cell carcinoma of the oral cavity primarily managed with ablative surgery.

Patients and Methods

A retrospective cohort study of patients with newly diagnosed oral squamous cell carcinoma from 2005 to 2019 was performed. Patients referred to and managed by a single surgeon (ERC) and who had not received any nutritional support in the preoperative period were included in the study. The primary predictor variable was preoperative albumin level. Other studied variables were patient demographic data and TNM stage. Complications related to primary ablative surgery represented the primary outcome variable. χ 2 analysis was completed to assess for significant associations between independent albumin groups (4+, 3.5 to 3.9, and 3.0 to 3.4 g/dL) in relation to postoperative complications. Multivariate logistic regression analysis was completed to control for clinical variables and medical comorbidities when testing the association between albumin and dehiscence.


The patient cohort included 268 individuals; of whom, 154 were men. The average age of the patients at surgery was 63 years. When controlling for all other variables, albumin was the only statistically significant predictor of postoperative dehiscence, P = .005. Patients with albumin of 3.5 to 3.9 g/dL had 3.24 times higher odds of dehiscence (95% confidence interval 1.42 to 7.38) in comparison with participants in the 4+ g/dL group. There was no difference of odds between the 3.0 to 3.4 group and the 4+ reference group.


Our study demonstrated that among those individuals meeting the inclusion criteria, there is a statistically significant association between lower albumin levels and postoperative complication rates, specifically dehiscence.

Malnutrition is commonly noted in patients with cancer. In particular, as many as 57% of patients diagnosed with head and neck cancer (HNC) will present with nutritional deficits. This observation can be explained in part by underlying comorbidities and risk factors for the development of both cancer and malnutrition, including tobacco abuse, alcohol abuse with liver disease, renal disease, diabetes, and cancer-induced cachexia in more advanced-staged disease. Malnutrition can occur secondary to difficulties with oral intake resulting from pain, dysphagia, and dysgeusia in patients with oral and oropharyngeal cancer. The neoplasm itself may also mechanically impede nutritional intake owing to trismus. Moreover, treatment for HNC often includes surgery, radiation therapy and chemotherapy; all of which may negatively contribute to malnutrition. Ablative surgery alters patient anatomy with resultant chewing and swallowing difficulty being observed. Finally, mucositis, dysgesia, anosmia, xerostomia, and nausea are all side effects of radiation therapy and chemotherapy and contribute to poor nutritional intake. It is therefore essential for patients with HNC to undergo consistent nutritional evaluation both preoperatively and postoperatively.

A number of biochemical markers can be used to indicate nutritional status such as albumin, prealbumin, and transferrin. Albumin is a prototypical indicator of both synthetic liver function and overall protein nutritional status. It represents the primary protein carrier while also preserving intravascular oncotic pressure and volume. Perioperative protein status has been shown to predict wound healing and other surgical complications in various clinical scenarios. In addition, as albumin has a longer half-life as compared with other acute phase proteins such as transferrin, prealbumin, and C-reactive protein, its value should be less affected by minor non-nutritional preoperative issues such as inflammatory- or cancer-related changes. , Albumin continues to be the defining indicator in starvation and has been noted as the best serum marker to correlate with hospitalized patient mortality.

Given that albumin is a marker of overall nutritional status, many authors have explored the relationship between perioperative albumin and outcomes in HNC with mixed results in both nonsurgical and surgical management. In patients with nonresectable pharyngeal and laryngeal cancer, hypoalbuminemia has been associated with inability to complete treatment and increased treatment failure. If salvage neck dissection is required after definitive radiotherapy or chemoradiotherapy, there also appears to be a trend toward a larger number of complications with lower albumin levels. Others, however, have found fistula formation after laryngectomy or hypopharyngectomy to be independent of albumin levels. Patients with primary surgical HNC with preoperative and/or postoperative hypoalbuminemia have been reported to experience wound dehiscence and/or have surgical site infections more frequently. , At least 1 study has reported postoperative complications occurring exclusively in hypoalbuminemic patients. It should be noted, however, that logistic or multiple regression analysis is not always performed to control for various confounders of wound healing. In addition, these studies variously include oral squamous cell carcinoma (OSCC), oropharyngeal squamous cell carcinoma, and other squamous and non–squamous cell HNC in their patient populations, and patients with and without preoperative radiotherapy are often included.

Abnormal albumin or prealbumin have also been evaluated in the specific setting of microvascular free-flap reconstruction of head and neck oncologic defects with mixed results. Low prealbumin (<10 mg/dL) has been found to predict free-flap failure but unfortunately is complicated by having multiple confounders and is not routinely measured. In addition to patients with diabetes or those who received neoadjuvant radiotherapy, individuals with a low body mass index or low albumin carried a greater risk of postoperative complications. Preoperative and/or postoperative hypoalbuminemia has variously been associated with free-flap failure. However, many patients included in previous studies are severely nutritionally compromised at baseline with albumin <3.0 g/dL.

The purpose of this study was to evaluate the association between preoperative albumin and postoperative complications in otherwise healthy patients presenting with a new diagnosis of OSCC who were treated with primary surgery. We hypothesized that an association would exist between preoperative hypoalbuminemia and postoperative complications. Our secondary goal was to investigate the association between demographic factors, medical comorbidities, and the TNM staging system and postoperative complications. Knowledge of the positive predictive value of preoperative hypoalbuminemia for postoperative complications would equip the surgeon with the ability to better inform the patient regarding the likely postoperative course.

Patients and Methods

Study Design and Sample

We designed and implemented a retrospective cohort study of patients treated for a diagnosis of OSCC by a single surgeon (ERC) between 2005 and 2019. Medical record numbers were retrieved for all patients with an ICD-10 diagnosis consistent with any anatomical subsite of OSCC treated at our institution during this inclusive 15-year period of time. The inclusion criterion was patients not receiving nutritional support in the preoperative period. The primary predictor variable was patients with squamous cell carcinoma of the oral cavity. The primary outcome variable was complications related to primary ablative surgery.

Applied exclusion criteria were 1) treatment by another surgeon, 2) carcinoma in situ or dysplasia, 3) the administration of neoadjuvant chemotherapy, 4) prior head and neck irradiation, 5) no documented albumin within 30 days before surgery, 6) less than 3-month follow-up, 7) patients who did not undergo primary surgery, and 8) patients observing chronic corticosteroid therapy.

All patients meeting inclusion and exclusion criteria underwent resection of their primary OSCC and simultaneous neck dissection. This study was approved by the Institutional Review Board at the University of Tennessee Medical Center Graduate School of Medicine (Knoxville, TN; IRB number 4593).

Study Variables

After obtaining qualifying medical record numbers as described previously, a chart review was performed for each patient. All preoperative consultation and clinic notes, laboratory data, operative reports, and histopathology reports were reviewed for each patient. Demographic, descriptive, and laboratory data were obtained including age at time of surgery, gender, tobacco use (former, current, or never), daily alcohol use (former, current, or never), history of hypertension and/or hyperlipidemia, history of diabetes, obesity, and history of liver disease. Laboratory data included preoperative serum albumin level within 30 days of surgery (categorized as 4+, 3.5 to 3.9, and 3.0 to 3.4 g/dL). Postoperative complications were recorded as infection requiring antibiotics only, infection requiring surgical debridement, wound opening exposing submucosal or deeper tissues without infection (dehiscence), hematoma, and seroma. A composite complications outcome was created by aggregating the 5 postoperative outcomes into a binary categorical outcome denoting no complication (coded as “0”) or presence of any type of outcome (coded as “1”). Clinical and pathologic TNM stage was also recorded for each patient. The study period involved the transition from the seventh to eighth edition of the American Joint Committee on Cancer TNM staging system for oral cavity squamous cell carcinoma such that depth of invasion and extranodal extension were reevaluated in each case and staging was designated as per the eighth edition.

Surgical Protocol

All patients underwent wide local excision of their primary cancers while observing 1- to 1.5-cm linear margins and ipsilateral or bilateral neck dissections as deemed appropriate by the senior surgeon. Patients were managed with selective neck dissections (I to III) when the neck examination was classified as N0 or type I modified radical neck dissection when the neck was classified as N+. Ipsilateral neck dissections were executed when the cancer was located greater than 1 cm from the midline, whereas bilateral neck dissections were executed when the cancer was located within 1 cm of the anatomical midline.

Histopathologic Analysis

The primary surgeon and pathology team met in the operating room to discuss the specimen including orientation after specimen removal. All specimens were transported directly from the operating room to the pathology laboratory and fixed in formalin. Processing included routine paraffin embedding, microtome slicing, and hematoxylin and eosin staining. Paraffin blocks were created for each centimeter of tumor specimen, and a microtome was used to cut tissue slices 2 to 6 micrometers in thickness from each block for histopathologic analysis. Slices were routinely prepared every 0.5 millimeter within each block.

Statistical Analysis

Frequency and percentage statistics were used to describe the clinical characteristics of the sample. χ 2 analysis was performed to test for any significant associations between the independent albumin groups (4+, 3.5 to 3.9, and 3.0 to 3.4 g/dL) in relation to postoperative complications, as well as demographic and clinical parameters of interest. Continuous outcomes such as age and weight were compared between the albumin groups using 1-way analysis of variance. Means and standard deviations were reported for the analyses of variance. Multivariate logistic regression analysis was performed to control for clinical variables and medical comorbidities when testing the association between albumin and dehiscence. Adjusted odds ratios with 95% confidence interval were reported and interpreted for the logistic regression analysis. Statistical significance was assumed at a P value of 0.05, and all analyses were performed using SPSS, version 26 (IBM Corp., Armonk, NY).

You're Reading a Preview

Become a DentistryKey membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here