Perioperative evaluation of glycaemic status in neck dissection: a retrospective analysis at a single hospital centre

Abstract

Diabetes mellitus is generally considered a risk factor for impaired wound healing. This study aimed to evaluate the glycaemic status of patients undergoing neck dissection and describe its impact on postoperative outcomes, especially wound healing. A retrospective analysis was performed of the preoperative, intraoperative, and postoperative glycaemic data obtained from the medical charts of 60 adult patients who had undergone 64 neck dissections. Nine of the 64 procedures were performed in diabetic patients (14.1%). The average glucose values were: preoperative 5.99 ± 1.25 mmol/l, intraoperative 8.90 ± 2.62 mmol/l, and postoperative 10.01 ± 2.49 mmol/l. All registered preoperative hyperglycaemia cases (eight cases) were diabetic. Postoperative insulin therapy was done in 14 procedures (21.9%). Wound healing complications were found in five patients (7.8%); there was no wound infection. There was no association of wound healing complications with preoperative diabetic status ( P = 1.000), preoperative glucose control ( P = 1.000), preoperative ( P = 0.469), intraoperative ( P = 0.248), and postoperative ( P = 0.158) glucose values, or with postoperative glucose control ( P = 0.577). These data do not support the association of stress-induced hyperglycaemia or diabetes mellitus with postoperative wound healing problems in neck dissection.

Diabetes mellitus is generally considered a risk factor for impaired wound healing. This negative impact is particularly well evidenced in chronic conditions such as leg and foot ulcers, in which factors that are both extrinsic and intrinsic to the wound and its biology play an important role. In spite of a strong physiological rationale, evidence showing diabetes mellitus to be a risk factor for impaired wound healing of acute surgical wounds is not so strongly supportive. Moreover, it appears that postoperative stress-induced hyperglycaemia, and not diabetic hyperglycaemia, may be a more important risk factor for surgical site infections that complicate surgical wound healing.

The neck dissection that accompanies resection of the primary lesion in the treatment of malignant tumours of the upper aerodigestive tract may also be complicated by surgical site infection wound complications. Wound complications in this location tend to result in fatal outcomes. However, the value of monitoring and treating hyperglycaemia, whether diabetic or stress-induced, in relation to the wound healing process in patients undergoing neck dissection, has not been established.

The aim of this study was to evaluate the glycaemic status of patients undergoing neck dissection and describe its impact on postoperative outcomes, predominantly wound healing.

Patients and methods

Between July 2007 and July 2011, 78 adult patients underwent 83 uncontaminated neck dissections at the study institution. A retrospective review of the medical charts was done, and the following data were collected: patient characteristics including age, gender, preoperative American Society of Anesthesiologists (ASA) status, diabetes status (non-diabetic, insulin-dependent diabetes mellitus (IDDM), and non-insulin-dependent diabetes mellitus (NIDDM)), and primary tumour location; preoperative glucose values and preoperative glucose control; operative details including the type of surgery (i.e. neck dissection), duration of surgery, duration of anaesthesia, and intraoperative glucose values; and postoperative data including postoperative glucose values, postoperative glucose control, wound healing complications, wound infections, time to drain removal, duration of intensive care unit (ICU) stay, and duration of hospital stay. All patients who underwent neck dissections during the study period (July 2007–July 2011) were included. We excluded all those with data (items listed above) missing from their medical charts.

General anaesthesia was induced and maintained using the same agents in all patients. In accordance with the local routine protocol, the same antibiotics (cefazolin 1 g and metronidazole 500 mg) were applied intravenously during anaesthesia induction, with the same doses repeated 3 h after the start of surgery. At the beginning of surgery an intravenous bolus of dexamethasone 16 mg was applied; this was not repeated.

Hyperglycaemia that required treatment was defined as a glucose level >10 mmol/l. Hypoglycaemia was defined as a glucose value <3.3 mmol/l. The type of surgery, i.e. neck dissection, was categorized as radical, modified radical, or selective according to standard definitions. A radical neck dissection involves the en-bloc removal of lymph node-bearing tissue on one side of the neck; the spinal accessory nerve, internal jugular vein, and sternocleidomastoid muscle are included in the specimen. A modified neck dissection comprises a radical neck dissection with preservation of one or more of the critical structures: the spinal accessory nerve, internal jugular vein, and sternocleidomastoid muscle. A selective neck dissection consists of the removal of only the lymph node groups; the spinal accessory nerve, internal jugular vein, and sternocleidomastoid muscle are spared. Any delay in wound healing, including a minor delay or a major delay with the development of a fistula, with or without evidence of infection, was defined as a wound healing complication. A wound infection was noted only for those with microbiology findings recorded in the medical charts. The final analysis included 60 patients who had undergone 64 procedures.

Statistical analysis

The statistical analysis was performed with SPSS software for Windows, version 9.0 (SPSS Inc., Chicago, IL, USA). Categorical data (gender, ASA score, primary tumour location, type of surgery, type of dissection, diabetes status) are expressed as frequencies and presented as numbers. The distribution of numerical data was determined with the Kolmogorov–Smirnov test and the Shapiro–Wilk test of normality. Normally distributed data are expressed as the mean ± standard deviation (SD), whereas non-normally distributed data are presented as the median with the interquartile range (IQR). The χ 2 test and Fisher’s exact test were used to compare categorical values between patients with and without wound complications. The unpaired t -test was used to compare continuous variables. Rank correlation analysis was performed to correlate preoperative, intraoperative, and postoperative glucose values with age, duration of surgery and anaesthesia, length of stay in the ICU and hospital, and time to drain removal. A P -value of <0.05 was considered statistically significant.

Results

Table 1 shows the demographic and clinical data of the study patients. Table 2 shows the glycaemic status data for the patients with head and neck cancer undergoing the different types of neck dissection.

Table 1
Patient demographic and clinical data ( n = 64).
Variables Results
Gender, n (%)
Male 45 (70.3%)
Female 19 (29.7%)
Age, years, mean ± SD 59.7 ± 12.93
ASA score, n (%)
1 6 (9.4%)
2 36 (56.3%)
3 22 (34.4%)
Primary tumour site, n (%)
Larynx 26 (40.6%)
Thyroid 12 (18.8%)
Pharynx 10 (15.6%)
Oral cavity 9 (14.1%)
Salivary glands 3 (4.7%)
Skin 1 (1.6%)
Not known 3 (4.7%)
Type of surgery, n (%)
Primary tumour extraction + neck dissection 46 (71.9%)
Neck dissection only 18 (28.1%)
Type of neck dissection, n (%)
Radical a 0 (0)
Modified radical b 45 (70.3%)
Selective c 19 (29.7%)
Duration of surgery, min, mean ± SD 318.2 ± 132.51
Duration of anaesthesia, min, mean ± SD 347.6 ± 134.84
Postoperative ICU stay
Yes 41 (64.1%)
No 23 (35.9%)
Wound healing complications 5 (7.8%)
Wound infection 0 (0)
Time to drain removal, days, mean ± SD 6.15 ± 2.70
ICU stay, days, median (IQR) 1 (0–11)
Hospital stay, days, median (IQR) 17 (6–161)
ASA score, American Society of Anaesthesiologists; ICU, intensive care unit; IQR, interquartile range; SD, standard deviation.

a Radical neck dissection involves en-bloc removal of lymph node-bearing tissue on one side of the neck (included in a specimen are the spinal accessory nerve, internal jugular vein, and sternocleidomastoid muscle).

b Modified radical dissection involves preservation of one or more of the following structures: spinal accessory nerve, internal jugular vein, sternocleidomastoid muscle.

c Selective dissection involves removal of only the lymph node groups.

Table 2
Perioperative glycaemic status of the study patients ( n = 64).
Variables Results
Preoperative diabetic status
NIDDM 7 (10.9%)
IDDM 2 (3.1%)
Preoperative glucose, mmol/l, mean ± SD 5.99 ± 1.25
Preoperative glucose control
Hypoglycaemia 0 (0)
Hyperglycaemia 8 (12.5%)
Intraoperative glucose, mmol/l, mean ± SD 8.90 ± 2.62
Intraoperative glucose control
Hypoglycaemia 0 (0)
Hyperglycaemia 0 (0)
Postoperative glucose, mmol/l, mean ± SD 10.01 ± 2.49
Postoperative glucose control
Hypoglycaemia 0 (0)
Hyperglycaemia 14 (21.9%)
IDDM, insulin-dependent diabetes mellitus; NIDDM, non-insulin-dependent diabetes mellitus; SD, standard deviation.

There was no case of registered hypoglycaemia in any patient prior to, during, or after surgery. However, hyperglycaemia was registered in 22 cases, and all registered cases of hyperglycaemia, no matter whether stress-induced or diabetic, were treated with short-acting insulin, in accordance with the conservative glucose control protocol. All registered preoperative hyperglycaemia cases (eight cases) were diabetic. There was no registered hyperglycaemia intraoperatively, so no treatment was given. Postoperatively, hyperglycaemia was treated in 14 cases (21.9%); four were diabetics and the remaining 10 were non-diabetics ( P = 0.096).

In all studied patients, the incidence of wound healing complications with pharyngocutaneous fistula formation was low (7.8%). There was no association between any of the patient or surgery characteristics and wound healing complications ( Table 3 ). In addition, there was no association between wound healing complications and preoperative diabetic status ( P = 1.000), preoperative glucose control ( P = 1.000), preoperative glucose value ( P = 0.469), intraoperative glucose value ( P = 0.248), postoperative glucose value ( P = 0.158), or postoperative glucose control ( P = 0.577) ( Table 3 ). For those with wound healing complications, there was no confirmed infection and no microorganisms were cultured other than those of the physiological flora.

Table 3
Perioperative characteristics and glycaemic status of the study patients and wound healing complications ( n = 64).
Variable Patients With Wound Healing Complications ( n = 5) Patients Without Wound Healing Complications ( n = 59) P -value
Gender, n (%) 0.310
Male 5 (100%) 40 (67.8%)
Female 0 (0) 19 (32.2%)
Age, years, mean ± SD 55.00 ± 7.35 60.10 ± 13.25 0.215
ASA score, n (%) 0.228
1 1 (20%) 5 (8.5%)
2 1 (20%) 35 (59.3%)
3 3 (60%) 19 (32.2%)
Primary tumour site, n (%) 0.889
Larynx 3 (60%) 23 (39.0%)
Thyroid 0 (0%) 12 (20.3%)
Pharynx 1 (20%) 9 (15.3%)
Oral cavity 1 (20%) 8 (13.6%)
Salivary glands 0 (0%) 3 (5.1%)
Skin 0 (0%) 1 (1.7%)
Type of surgery, n (%) 1.000
Primary tumour extraction + neck dissection 4 (80%) 42 (71.2%)
Neck dissection only 1 (20%) 17 (28.8%)
Type of neck dissection, n (%) 1.000
Radical a 0 (0) 0 (0)
Modified radical b 4 (80%) 41 (69.5%)
Selective c 1 (20%) 18 (30.5%)
Duration of surgery, min, mean ± SD 349 ± 107.67 315.56 ± 134.85 0.542
Duration of anaesthesia, min, mean ± SD 371 ± 64.65 345 ± 139.31 0.479
Postoperative ICU stay, n (%) 0.150
Yes 5 (100%) 36 (61.0%)
No 0 (0) 23 (39.0%)
Wound infection, n (%) 0 (0%) 0 (0%) NA
Time to drain removal, days, mean ± SD 5.20 ± 2.17 6.24 ± 2.74 0.360
ICU stay, days, median (IQR) 1 (1–1) 1 (0–11) 0.882
Hospital stay, days, median (IQR) 28 (23–43) 16 (6–161) 0.126
Preoperative diabetic status, n (%) 0 (0%) 9 (15.3%) 1.000
Preoperative glucose, mmol/l, mean ± SD 5.74 ± 0.70 6.01 ± 1.29 0.469
Preoperative glucose control, n (%) 0 (0%) 8 (13.6%) 1.000
Intraoperative glucose, mmol/l, mean ± SD 7.70 ± 0.99 9.04 ± 2.73 0.248
Postoperative glucose, mmol/l, mean ± SD 9.12 ± 1.91 11.69 ± 10.39 0.158
Postoperative glucose control, n (%) 0 (0%) 14 (23.7%) 0.577
ASA score, American Society of Anaesthesiologists; ICU, intensive care unit; IQR, interquartile range; NA, not applicable; SD, standard deviation.

a Radical neck dissection involves en-bloc removal of lymph node-bearing tissue on one side of the neck (included in a specimen are the spinal accessory nerve, internal jugular vein, and sternocleidomastoid muscle).

b Modified radical dissection involves preservation of one or more of the following structures: spinal accessory nerve, internal jugular vein, sternocleidomastoid muscle.

c Selective dissection involves removal of only the lymph node groups.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free dental videos. Join our Telegram channel

Jan 19, 2018 | Posted by in Oral and Maxillofacial Surgery | Comments Off on Perioperative evaluation of glycaemic status in neck dissection: a retrospective analysis at a single hospital centre

VIDEdental - Online dental courses

Get VIDEdental app for watching clinical videos