Factors Affecting Postoperative Complications After Reconstructive Surgery in Oral Carcinoma Patients: A Prospective Study of 100 Patients

Abstract

Postoperative complications may result in significant functional morbidity, poor cosmetic results, prolonged hospitalization, preclusion of optimal treatment for the cancer, or even be pose threat to life. We prospectively assessed postoperative complications in 100 patients who underwent surgical resection with free or pedicled flap reconstruction as a primary modality of treatment in patients with carcinoma of the oral cavity. One hundred consecutive patients who underwent reconstructive surgery for oral cancer were prospectively analyzed for age, gender, comorbidities, tumor stage, nodal stage, primary sub-site of tumour, reconstructive procedure (free or pedicled), duration of surgery, blood transfusions during surgery, preoperative weight and body mass index, patient generated subjective global assessment status and tracheostomy to determine their effect on postoperative complications as determined on the CD scale. The sample comprised 100 patients with a mean age of 52.12 years (range 24–80 years) and 74% men (M:F ratio 3:1). A total of 40 patients developed surgical complications (including two deaths) while medical complications were seen in 10 patients (including one death). Tracheostomy (52 vs. 7%, p = 0.002) and age (54 vs. 49 years, p = 0.031) were associated with higher complication rate. Higher age and tracheostomy is associated with higher complications in postoperative period.

Introduction

Squamous cell carcinoma of the oral cavity (OSCC) is most common cancer in South Central Asia, including India [1]. Surgery remains the primary modality of treatment and reconstructive surgery is often necessitated for restoration of form and vital functions like speech, swallowing and respiration. Reconstructive surgery in the form of free or pedicled flaps has become the standard method of reconstruction of complex defects in the head and neck [2]. While reconstructive surgery improves functional and aesthetic outcomes and improves quality of life [3], these procedures add to operating time, donor site morbidities, requirement of blood and blood products and adjunctive procedures like tracheostomy. The impact of reconstructive surgery on the spectrum of post-operative complications has not been investigated comprehensively in literature.

Postoperative complications may result in significant functional morbidity, poor cosmetic results, prolonged hospitalization, preclusion of optimal treatment for the cancer, or even be life-threatening, resulting in compromised postoperative patient care [4].

Postoperative complications are generally segregated into surgical and medical complications. Although it is easy to differentiate between these two types of complications, it is often difficult to quantify them for purpose of evaluation. Considerable heterogeneity and inconsistency exists in present literature as a result of subjective interpretation and non-standard reporting of complications, stemming from shortcomings such as the absence or lack of consensus about definitions of complications; the absence of, or varying definitions of, the postoperative period investigated; or the lack of accurate grading of complications [5–7].

The Clavien–Dindo Classification is five scale classification provides an objective, simple, reliable, and reproducible way of reporting negative events after surgery (Table 1) [8]. It has been used for standardizing complications after free flap resections in head and neck cancers but there is paucity of studies in head and neck surgery assessing the utility of the Clavien–Dindo scale in head and neck cancers [9–12].

Table 1.

Clavien–Dindo classification

Grade	Description
I	Any deviation from the normal postoperative course without the need for pharmacological treatment or surgical, endoscopic, and radiological interventions Allowed therapeutic regimens are: drugs as antiemetics, antipyretics, analgetics, diuretics, electrolytes, and physiotherapy. This grade also includes wound infections opened at the bedside
II	Requiring pharmacological treatment with drugs other than such allowed for grade I complications Blood transfusions and total parenteral nutrition are also included
III	Requiring surgical, endoscopic or radiological intervention Grade IIIa Intervention not under general anesthesia Grade IIIb Intervention under general anesthesia
IV	Life-threatening complication (including CNS complications) requiring ICU management Grade IVa Single organ dysfunction (including dialysis) Grade IVb Multiorgan dysfunction
V	Death of a patient

Suffix “d” If the patient suffers from a complication at the time of discharge, the suffix “d” (for “disability”) is added to the respective grade of complication. This label indicates the need for a Follow-up to fully evaluate the complication

In this study, we prospectively assessed postoperative complications in 100 patients who underwent surgical resection with free or pedicled flap reconstruction as a primary modality of treatment in patients with carcinoma of the oral cavity.

Materials and Methods

This was a prospective cohort study and the study sample was derived from a population of 100 patients who underwent surgical resection with reconstruction for OSCC from February 2014 through September 2016 with prior approval from the Institutional Review Board. Patients eligible for study inclusion were previously untreated and underwent tumor resection with reconstruction as their primary method of treatment. Patients were excluded prior to their inclusion, if they had a recurrent tumour or resection was planned for primary closure without a reconstructive procedure.

Only complications that occurred during the postoperative period when patient was still in the hospital were considered for the study. Other study variables included age, gender, previous medical history, tumor stage (defined by size of tumor), nodal stage (AJCC 7), primary site of tumour, type of reconstructive procedure (free or pedicled), duration of surgery, blood transfusions during surgery, preoperative weight and body mass index, preoperative scored Patient Generated Subjective Global Assessment (PG-SGA) [13] and tracheostomy.

PG-SGA is a validated method of nutritional assessment, developed specifically for cancer patients, based on the features of a medical history (weight change, dietary intake change, gastrointestinal symptoms that have persisted for more than 2 weeks, changes in functional capacity), physical examination (loss of subcutaneous fat, muscle wasting, ankle/sacral edema and ascites), questions regarding presence of nutritional symptoms and short term weight loss [13, 14]. The scored PG-SGA is a further development of the PG-SGA concept that incorporates a numerical score as well as providing a global rating of well-nourished, moderately or suspected of being malnourished or severely malnourished [15]. Higher the score, greater is the risk for malnutrition. The scored PG-SGA has been found to be a quick, valid and reliable nutrition assessment tool that enables malnourished hospital patients with cancer to be identified and triaged for nutrition support. Its advantages include that the components of the medical history are completed by the patient which is less time intensive, it identifies a more extensive range of nutrition impact symptoms and the scoring system allows standardization in reporting the results [16]. The scored PG-SGA has been accepted by the Oncology Nutrition Dietetic Practice Group of the American Dietetic Association as the standard for nutrition assessment for patients with cancer.

The Clavien–Dindo Classification is used to report the complications in surgical patients. The complications are divided in five groups according to the severity [8]. Although extensively used in General Surgery, the literature for use of Clavien–Dindo scale in head and neck surgery is sparse. Monteiro et al. [9] applied the Clavien–Dindo scale to hypothetical cases of common head and neck surgical complications and reported that the inter-observer reliability for the scale was moderate to high. Although there were specific concerns regarding its utility in head and neck, overall, it was well received by the respondents. Also, the complication grade was related to the length of hospital stay. McMahon et al. [12] found it useful to report the complications in patients undergoing major head and neck surgery with free flap repair.

Descriptive and inferential statistics were performed with SPSS 17 (IBM Corp., Chicago, Illinois, USA). Categorical variables were analyzed using Chi square test, effect of continuous variables on postoperative complications was analyzed using independent samples t test. Both univariate and multivariate analyses were performed.

Observations and Results

In 100 patients who underwent surgery for OSCC, medical complications were observed in 11 patients, with one death due to medical cause. Surgical complications were seen in 40 patients with two deaths attributed to surgical causes. Wound infection was the most common complication, seen in 19 patients. Six patients had a complete flap loss, while five others had a partial flap loss. Orocutaneous fistula occurred in six patients and five patients had wound hematoma. Respiratory infections and pleural effusion were most common medical complications, seen in five patients. Details of the complications according to the Clavien–Dindo classification is depicted in Table 2.

Table 2.

Complications as per Clavien–Dindo grading

Complication	Clavien–Dindo grade							Total
	I	II	IIIa	IIIb	IVa	IVb	V
Wound infection	0	10	4	3	0	0	2	19
Partial flap loss	1	0	3	1	0	0	0	5
Complete flap loss	0	0	0	6	0	0	0	6
Flap detachment	1	1	1	0	0	0	0	3
Orocutaneous fistula	1	1	3	1	0	0	0	6
Wound hematoma	0	0	1	4	0	0	0	5
Serous collection	2	0	0	0	0	0	0	2
Wound dehiscence	1	0	1	0	0	0	0	2
Respiratory infection	1	3	1	0	0	0	0	5
Myocardial infarction/cardiac arrest	0	1	0	0	0	0	1	2
Urinary retention	1	0	0	0	0	0	0	1
Septicaemia	1	0	0	0	0	0	0	1
Psychological issues	1	0	0	0	0	0	0	1

Mean age of the patients was 52.12 years with patients ranging from 24 to 80 years. There were 74 males and 26 females amongst the patients.

Tracheostomy was associated with higher complications rate with 52% patients with tracheostomy having complications of any grade (p = 0.003) (Table 3).

Table 3.

Complications in different groups

Factor	N	Postoperative complications seen in (%)	p value
Gender
Male	74	30 (40.54)	0.065
Female	26	16 (61.53)
T Stage
T1	14	3 (21.42)	0.178
T2	27	13 (48.14)
T3	13	6 (46.15)
T4	46	24 (52.17)
N stage
N0	42	20 (47.61)	0.269
N1	26	9 (34.61)
N2a	2	2 (100)
N2b	22	10 (45.45)
N2c	7	5 (71.42)
N3	1	0 (0)
Pathological stage
Stage I	10	3 (30)	0.195
Stage II	13	8 (61.53)
Stege III	12	3 (25)
Stege IV	65	32 (49.23)
PG-SGA group
SGA A	60	26 (43.33)	0.282
SGA B	32	15 (46.88)
SGA C	02	2 (100)
Flap type
Free	88	42 (47.72)	0.474
Distant pedicled	07	3 (42.85)
Local pedicled	05	1 (20)
Intraoperative blood transfusion
None	70	29 (41.42)	0.066
One unit	14	5 (35.71)
Two units	14	10 (71.42)
More than two units	02	2 (100)
Tracheostomy
Done	86	45 (52.32)	0.002
Not done	14	1 (7.14)

There was trend towards higher complication rates in females (62 vs. 41%) and in patients with higher intraoperative blood loss requiring 2 or more units blood transfusion (75%) (Table 3).

Coexisting comorbidities were not seen to contribute to the incidence of postoperative complications (Table 4).

Table 4.

Effect of various factors on complications

Factor	Mean (95% CI)		p value
	Complications	No complications
Age (years)	54.27 (50.59–57.94)	49.47 (46.33–52.60)	0.031
Weight (kg)	64.18 (60.13–68.23)	67.95 (63.82–72.08)	0.233
BMI (kg/m2)	25.06 (23.61–26.51)	25.92 (24.62–27.21)	0.351
Global PG assessment	3.61 (2.81–4.41)	3.62 (3.02–4.21)	0.943
Duration of surgery (min)	580 (550–610)	560 (530–590)	0.336

Development off complications significantly affected the hospital stay of the patients (mean 13.88 vs 10.75 days, p = 0.010) (Table 5).

Table 5.

Effect of comorbidities on complications

Co-morbidity	Complications	No complications	p value
None	22	34	0.159
Cardiovascular	16	16	0.669
Respiratory	01	00	0.460
Endocrine	13	10	0.341
Tuberculosis	01	00	0.460
Surgical	02	03	1.000

Amongst the 100 patients, Free Fibula Flap was the most common Flap used, with average hospital stay of 13.12 days. The types of flaps used and the average hospital stay is depicted in Table 6.

Table 6.

Hospital stay according to the type of flap

Type of flap	Number of patients	Mean hospital stay (days)
Free flaps
Free fibula flap	31	13.12
Anterolateral thigh flap	22	15.04
Radial artery forearm flap	21	10.5
Lateral arm flap	11	11.45
Latissimus dorsi free flap	1	11
Groin free flap	1	9
Rectus abdominis flap	1	8
Pedicled flaps
Pectoralis major myocutaneous flap	7	9.85
Nasolabial flap	2	6.5
Temporalis flap	2	5
Tongue flap	1	10

Discussion

Several studies have attempted to identify patients at high risk for developing complications following head and neck oncologic procedures [17]. The factors identified previously include advanced stage of tumor, reconstruction with a myocutaneous flap, nutritional status, concomitant disease, preoperative radiotherapy, duration of surgery, classification of procedure, and antibiotic prophylaxis [18, 19].

Considerable heterogeneity and inconsistency exists in present literature as a result of subjective interpretation and non-standard reporting of complications, stemming from shortcomings such as the absence or lack of consensus about definitions of complications; the absence of, or varying definitions of, the postoperative period investigated; or the lack of accurate grading of complications [5–7]. We attempted to mitigate these lacunae and adopted the Clavien–Dindo classification system to record and grade postoperative complications in a group of patients having undergoing reconstructive procedure for OSCC with either pedicled or free flap at the same time as their oral cancer was resected. We have tried to minimize reporting errors and subjective interpretation of data about complications by defining both them and the period of morbidity investigated, and by using a standard grading system to rank them. The proposed morbidity scale provided an efficient and accurate format for assessing their incidence and severity. All the surgeries were clean contaminated surgeries and all patients received intravenous cefuroxime and metronidazole as prophylaxis, first dose given at the time of induction.

In our study, medical and surgical complications were seen in 11 and 40% patients respectively. Major medical complications were seen in 2 out of 11 (18%) (Clavien–Dindo grade 3 or more) and 25 out of 40 (63%) patients had major surgical complications, making 53% complications being categorized as major overall. Despite the high rates, mortality was low at 3%, in close agreement with Perisanidis et al. [20].

Our study showed that older patients have a higher incidence of postoperative complications. Complications were observed in 16 out of 26 females (61.53%, 14 surgical, 3 medical), of which 44% were major. 30 in 74 males faced complications (40.54%, 28 surgical, 8 medical), that comprised 66% major complications. The rate of complications, although more in females, they faced fewer major complications.

Malnutrition has long been suggested as predisposing to postoperative wound infections in the head and neck cancer patient. Protein-calorie malnutrition results in impaired wound healing, reduced immune-competence, and decreased tolerance to surgery [21]. Brown suggested that more accurate methods of nutritional assessment would demonstrate definite and significant correlation to wound infection [22]. We used weight, body mass index and PG-SGA status as the parameters to reflect on nutritional status of the patient. In contrast to previous studies, our study failed to show a relationship between malnutrition and postoperative complications in patients.

Free microvascular tissue transfer using free flaps are technically more complicated and preferred for complex defects of the oral cavity and require greater operating time. In spite of this, there was no difference in complications rates with different sub-sites of the tumour or when free or pedicled flaps were used for reconstruction reinforcing the primacy of free tissue transfer in the routine reconstructive effort.

Robbins et al. [4] showed that higher T and N stage is associated with higher wound infection rates. Others have previously noted a higher wound infection rate with stage III and IV disease. It is clear then that the more advanced the tumor, the greater the predisposition to wound infection. As Brown suggested, excision of extensive tumors leaves larger defects and more difficult reconstructions in which it is more likely to result in errors in surgical technique and judgement [22]. Our study failed to show any correlations between T and N stage and overall pathological stage and postoperative complications.

Several factors related to the surgical procedure have been associated with wound infection [17]. Duration of the surgery, requirement of blood products during surgery were used to adjudge the complexity of surgery in this study. These factors did not affect the occurrence of postoperative complications. Robbins et al. [4] linked complexity of the surgical procedure, as well as flap reconstruction to wound infection. Brown and Johnson, among others, have reported on the problems associated with more extensive surgery involving myocutaneous flaps. They suggested that the increased incidence of postoperative wound infection in those patients who undergo extensive procedures is likely the result of prolonged contamination of the wound by saliva which is in turn caused by inadequate closure at the time of surgery or ischemia and subsequent wound separation in the postoperative period. It implies that surgical error is a factor of procedure complexity. In fact, they postulated that surgical error may be the most important factor predisposing to infection in major head and neck procedures. This hypothesis would be difficult to test, as technique varies according to the procedure and the surgeon [22]. In our group, such differences were not evident as most patients (%) were homogenously of higher stages and all required complex reconstructive surgery where the effect of factors like blood transfusion, operating time would not have been evident.

The presence of a tracheostomy significantly affected the postoperative complications in this study. Of the 14 patients who did not have a tracheostomy, only 1 patient (7.14%) developed complications, which was Clavien–Dindo grade IIIa respiratory complication. 45 out of 86 patients in tracheostomy group developed complications (53.32%, 40 surgical, 10 medical). Of these patients, 26 (57.77%) had major complications. Similar findings were noted by Robbins et al. We believe that this could have been because tracheostomy was performed in patients with more complex surgeries and also the tracheostomy tube may have led to bacterial colonization of secretions and subsequent infections.

Our study did not show a relation between patient comorbidities and postoperative complications. This is in contrast with other studies, which have shown a relation between patients’ comorbidities and complete loss of a flap [20]. We assume that preoperative optimization and post-operative vigilance would have ameliorated the effect of these comorbid factors.

Postoperative complications significantly prolonged the hospital stay of the patients. Prolonged hospital stay increases the indirect and direct cost burden to the patient. Duration of hospital stay has been used in several studies as a surrogate for severity of complications [6, 7, 23]. However, duration of stay in hospital is strongly influenced by the medical policy of each individual department, and therefore cannot be considered a reliable criterion for comparing outcomes either nationally or internationally. The cost implications were beyond the purview of this study.

Some authors have reported that postoperative complications after oral and oropharyngeal cancer surgery have a deleterious impact on long-term outcome [24, 25]. One of the more interesting outcomes of these studies would have been the impact of postoperative complications on survival of the study population. In this study, the mean follow up of the patients is about 1 year which is too short to make any meaningful inferences. We are following these patients and will present those results at a later date.

Conclusion

The Clavien–Dindo classification can easily be used to evaluate postoperative complications after free flap surgery. Higher age and tracheostomy is associated with higher complications in postoperative period.

Compliance with Ethical Standards

Conflict of interest

All authors declare that they have no conflict of interest.