Prophylactic Glycopyrrolate Administration to Aid Otolaryngology Residents in Prevention of Post-laryngectomy Pharyngocutaneous Fistula

Abstract

Pharyngocutaneous fistula (PCF) after laryngectomy occurs when there is a failure in the pharyngeal repair resulting in a salivary leak (Dedivitis et al. in Acta Otorhinolaryngol Ital 27(1):2–5, 2007). In the post-operative period this complication is not only a challenge to the surgeon but also distressing to the patient and his family. We conducted a retrospective cum prospective study, to research the effect of starting prophylactic glycopyrrolate intravenous injection in total laryngectomy. Fifty patients underwent total laryngectomy with modified radical neck dissection over a period of 5 years by the trainee otolaryngologists. Of these 25 patients were started on glycopyrrolate for 7 post-operative days. And 25 control patients, who were not given glycopyrrolate. All the patients were observed for evidence of PCF. Of the total 50 patients, only 1 patient in study group and 12 patients in the control group developed PCF. This was statistically significant (p value 0.0039). Tumor stage, site, and gender were found to have no statistically significant correlation with development of PCF. The prophylactic administration of glycopyrrolate was found to reduce the odds of development of PCF in total laryngectomy cases. There was associated blurring of vision and constipation amongst the cases subjects, which was only transient with complete recovery on cessation of drug.

Introduction

Pharyngocutaneous fistula (PCF) is one of the common complications after laryngeal and pharyngeal radical surgery. It is a major cause of morbidity, impaired the quality of life and prolonged hospital stay for patients undergoing laryngectomy surgery. The reported incidence of PCF ranges from 7.6 to 50%. PCF is a demoralizing complication for the surgeons especially for the trainee otorhinolaryngology residents. Factors that lead to increased risk of PCF are pre-operative radiation, poor nutritional status, haemoglobin levels, blood transfusion, prolonged tracheostomy, surgical site infection, surgical technique of closure, neck dissection, surgeon’s experience. Until today only few studies have focused on prevention of PCF in post laryngectomy patients. It is believed that even local irritation by nasogastric tube (NGT), early feeding can increase the chances of PCF and hence many surgeons believe in keeping the patient nil per orally for 7 days [2]. Even when patient is nil per orally, the excessive salivation leads to initiation of swallowing and distension and movement of the neopharynx. Hence neopharynx is never really at rest. Reduction of salivation using anticholinergic drug such as glycopyrrolate may aid in better healing of the neopharynx. Incidence of PCF was fairly high in the hands of trainee residents in our setup, especially in the background of poor nutritional status of the patients. This study shows the advantage of glycopyrrolate on reducing the post-operative PCF in total laryngectomy patients.

Material and Methods

A retrospective cum prospective study of 50 patients (42 male and 8 female) with age ranging from 35 to 60 years, who underwent total laryngectomy procedures with modified neck dissection for malignant neoplasm of the larynx and hypopharynx was done. The study comprised of two groups, each group comprising of 25 patients. Test group (case) subjects were prospective cases, who were operated during time period of January 2018–December 2019 (2 years) for malignancy of larynx or hypopharynx in a Government tertiary hospital. The case subjects were started post-operatively on intravenous glycopyrrolate injection three times a day till 8th post-operative day. Control subjects were age, gender, smoking exposure, and surgical procedure matched. Control subjects were selected from the retrospective record data, in whom prophylactic glycopyrrolate post operatively was not given. Retrospective data was collected for cases operated during the years 2015–2018.

The inclusion criteria for Case group were as follows:

1. Malignancies of larynx and hypopharynx who have consented for the study.

2. T stage-T3 or T4, N1, M0

3. Patient who underwent total laryngectomy with modified neck dissection.

4. Patients in whom post operatively sodium glycopyrrolate was given till 8th post-operative day.

5. Cases operated by the residents of otolaryngology and head and neck surgery under supervision of mentors.

6. Normal preoperative haemoglobin, albumin, and nutritional status.

7. Pharyngeal mucosa more than 2.5 cm for closure.

8. Good family and emotional support in the post-operative period.

The inclusion criteria for the controls were as follows:

1. Age, gender, smoking exposure matched control subjects

2. Malignancies of larynx and hypopharynx

3. T stage-T3 or T4, N1, M0

4. Patient who underwent total laryngectomy with modified neck dissection

5. Patients in whom post operatively sodium glycopyrrolate was not given post operatively.

6. Cases operated by the residents of otolaryngology and head and neck surgery under supervision of mentors

7. Normal preoperative haemoglobin, albumin, and nutritional status.

8. Pharyngeal mucosa more than 2.5 cm for closure.

9. Good family and emotional support in the post-operative period.

The exclusion criteria for all patients in both the study and control groups were as follows.

1. History of pre-operative radiotherapy

2. Diabetes, poor nutritional status (assessed by body weight, blood levels of haemoglobin, albumin, globulin and total proteins)

3. Wide pharyngeal involvement in which mucosa left behind for primary closure was less than 2.5 cm.

4. Those who needed primary reconstruction technique such as axial flaps or free flaps.

5. Patients who underwent more radical procedure with or without reconstruction.

6. Primary tracheo-oesophageal puncture

7. Systemic complications in the post-operative period e.g. cardiac complications or chest complications.

8. Patients diagnosed with glaucoma and prostatomegaly.

9. Pre-operatively tracheotomised for more than 14 days.

All subjects underwent detailed clinical examinations and investigations included a routine blood profile, serum albumin and globulins, computed tomography scans and chest radiographs as a part of preoperative evaluation. Majority of the patients were from poor socioeconomic background with poor nutritional status. Preoperatively all the subjects underwent nutritional rehabilitation and only those with normal serum albumin and haemoglobin were subjected to the proposed surgery. All patients were assessed by panendoscopy and biopsy. All patients were operated under general anaesthesia. A Gluck-Sorenson U shaped incision was used for laryngectomies with extension for modified radical neck dissection (Fig. 1). During the laryngeal release, the inferior constrictor muscles were separated from the thyroid ala (Fig. 2). We can visualise the tumor on right supraglottis after the entry into pharynx. The pharyngeal defect was closed with 3–0 vicryl sutures after introduction of NGT. The pharynx was closed in 3 layers in I shape with interrupted sutures without posterior myotomy. The first layer included the sub mucosa in an edge inverting manner, the second layer included sutures taken adjacent to the first layer and the third layer consisting of the pharyngeal constrictors. The tracheostoma was fashioned by suturing upper and lower skin flaps. The skin was closed in two layers, and drains were kept in place for 2 post-operative days. Patients were put on third generation cephalosporin (ceftriaxone sodium) and metronidazole for one week postoperatively.

Fig. 1.

Showing flap raised using Gluck sorenson’s incision

Fig. 2.

Showing supraglottic groth seen after pharynx entry

All patients were put on anti-reflux measure ranitidine hydrochloride given intravenously. In the study group glycopyrrolate injection 0.2 mg thrice daily was given intravenously for 8 days post operatively. All patients were put on head end elevation position post operatively.

All patients were given clear fluids orally on 8th post-operative day and were observed for leaks. In the absence of leak, the NGT was removed and liquid feeds were given for 2 days followed by semisolid feeds (soft diets such as bread, boiled rice, porridge etc.) and the patients were discharged once they were able to maintain adequate nutrition orally.

In case of leaks, NGT was kept in situ and pressure dressing was applied to reduce leak, glycopyrrolate was continued in study group. All the patients in both groups were strictly observed for PCF and oral feeds were stopped if the fistula developed.

Chi square test, odds ratio, and fisher exact test were used for analysis of data and p value of < 0.05 was considered significant in cases where the numbers were zero.

Results

The study consisted of a total of 50 patients. Study group comprising of 25 patients who were started on 0.2 mg intravenous glycopyrrolate post operatively thrice a day for 8 post-operative days and control group comprising of 25 patients in whom post-operative glycopyrrolate injections were not administered.

Among the 50 patients, 30 patients had malignancies of larynx (15 each in case and control) and 20 patients had malignancies of hypopharynx (10 each in case and control) (Table 1). All the patients underwent total laryngectomy with ipsilateral modified neck dissection with primary pharyngeal closure done horizontally. All patients were started on oral feeds on 8th postoperative day and observed for any evidence of leak. One patient in study group and 12 patients in control group developed PCF (Table 2).

Table 1.

Tumor distribution based on site in case and control groups (N = 50)

Site	Case (n = 25)	Control (n = 25)
Larynx	s15	15
Hypopharynx	10	10

Table 2.

Association of use of Glycopyrrolate and PCF occurrence (N = 50)

Glycopyrrolate use	Pharyngocutaneous fistula
	Yes	No
Yes	1	24
No	12	13

Chi square T test done. p value 0.0039 (significant)

Among the cases who developed PCF, majority improved with conservative management consisting of NGT feeds and local wound dressing. Four patients required debridement and secondary suturing. One patient required pectoralis major myocutaneous flap reconstruction (Fig. 3).

Fig. 3.

Patients with flap necrosis secondary to PCF

In all the cases the histopathological assessment of surgical specimen was done. And all patients were confirmed to have Squamous cell carcinoma.

Our study showed a significant correlation between prophylactic use of glycopyrrolate and reduced occurrence of PCF in study group when compared with control group (chi square test p value 0.0039) (Table 2). We also found that the risk of occurrence of PCF in case group was 4% as compared to control group which was 48%. The odds ratio was calculated, which was 0.045. This signifies that the use of prophylactic sodium glycopyrrolate lowered the odds of development of PCF among the cases of total laryngectomy and modified radical neck dissection.

We have found no significant correlation between age, gender, stage, and site of tumor and the occurrence of PCF (fisher exact test p value 0.46) (Table 3). The complications in the post-operative period are shown in Table 4.

Table 3.

Occurrence of use of PCF with respect to site of tumor (N = 50)

Site	Case group (n = 25)	Control group (n = 25)
Larynx	1	5
Hypopharynx	0	7

Fisher exact test done. p value 0.46 (not significant)

Table 4.

Complications in post operative period (N = 50)

Complications	Glycopyrrolate (n = 25)	Control (n = 25)
Wound infections	2 (8.0%)	2 (8.0%)
Flap necrosis	0 (0.0%)	2 (8.0%)
Surgical emphysema	0 (0.0%)	0 (0.0%)
Carotid blowout	0 (0.0%)	0 (0.0%)
Aspiration	1 (4.0%)	1 (4.0%)
Rentention of urine	2 (8.0%)	1 (4.0%)
Pharyngocutaneous fistula	1 (4.0%)	12 (48.0%)
Dryness of mouth	5 (20.0%)	0 (0.0%)
Blurring of vision	19 (76.0%)	0 (0.0%)
Constipation	5 (20.0%)	2 (8.0%)

We noticed that in the case group, in whom sodium glycopyrrolate was prophylactically used, 19 subjects complained of transient blurring of vision and 5 subjects complained of constipation during the course of treatment. Whereas none of the patient in control group reported of any of these symptoms. None of the case group subjects reported of long term visual disturbance during the follow up.

Discussion

In 1873 Theodore Billroth carried out first laryngectomy for cancer patient [3], since then PCF has been a dreaded complication. Inability to start oral feeds and PCF is demoralizing for the surgeon as well as the patient who already has lost voice. It also increases expense, and delays starting adjuvant radiation therapy [4]. The incidence of PCF reported in literature is around 9–25% [5]. Though the incidence of fistula has been published in many papers, the literature is sparse on prevention of PCF.

According to Cantrell the factors that influence fistula formation are poor nutrition, pre-operative irradiation, pre-operative tracheostomy, larger tumors that requires larger resection of pharyngeal mucosa, concurrent radical neck dissection, type of incision, improper closure of pharyngeal mucosa, closing wound under tension, failure to drain the wound, failure to use an NGT, postoperative infections, low post-operative hemoglobin levels and feeding the patient too soon [6]. A meta-analysis of post-laryngectomy PCF showed that pre-operative radiotherapy increased the risk of PCF formation, and the severity and duration of fistula were greater than those without pre-operative radiotherapy [7]. Nevertheless its prevention remains very difficult.

For many years delaying pharyngeal feeds for more than a week after laryngectomy was considered as a way of preventing leak. Several authors have cited to support the notion that delayed oral feedings up to 3 weeks lessens the incidence of fistula [8]. Some authors suggest starting oral feeds as early as second postoperative day. Boyce and Meyers [8] conducted a poll that revealed 84.5% of practicing head and neck surgeons wait at least 7 days to feed their patients after laryngectomy and only 2.8% feed at 4 days or before. Due to lack of randomized control trials none of the studies have validated the hypothesis. According to Boyce and Meyers early oral feeds should be considered as it may lead to shortened hospital stay, cost saving and improved patient comfort [2]. In our study we have started oral feeds on 8th post-operative day in order to provide adequate time for healing. Early feeding and pharyngeal movement may prevent strengthening of weak mucosal closure.

Surgical technique is considered as most important risk factor for development of PCF. Stell and Maran [9] have advocated vertical pharyngeal closure. Also according to Stell and Cooney meticulous suturing is important for reducing PCF. Levent in his study found PCF rate was significantly less when vicryl was used for pharyngeal closure compared with catgut. This is because the vicryl is much stronger, is less inflammatory, and has longer half-life [10]. We have previously described our technique for pharyngeal closure.

Many authors believe that movement of NGT to be irritating to the suture line. The presence of the NGT is risk factor for gastroesophageal reflux which may cause irritation of the wound. Shah and Ingle used soft NGT for reduction of pressure on the suture line and thus preventing PCF [11]. All our patients were put on 18 FG NGT and started on anti-reflux medications.

Factors that promote wound healing are good broad spectrum antibiotics which prevents perioperative and immediate postoperative wound infections and wound break down [12]. A good dressing with a proper drain at wound site prevents dead space under the skin flap, thereby preventing the accumulation of the blood and secretions under the skin flap and deeper tissues. Nutrition with high protein diet, adequate fluids, and multivitamins promotes better wound healing.

Our concept of starting the glycopyrrolate injection is based on following considerations, swallowing and feeding stresses the suture line in the neopharynx [13]. Saliva also has a role to play in PCF, excessive production of saliva initiates swallowing movements [14]. This could put pressure on the pharyngeal suture line which can result in give way and leak.

The fact that the patient is swallowing about 1.5 L of saliva over 24 h from the moment of awakening from the operation supports its role in initiating the anastomotic leak. Vaghela et al. [15] in their study showed that with administration of glycopyrrolate in patients of anastomotic leak after primary repair of oesophageal atresia there was reduced oral secretions, which in turn helped with healing of the anastomotic dehiscence.

In our study we have used glycopyrrolate which is rapidly absorbed (1–2 min) after intravenous injection. Glycopyrrolate binds competitively to the muscarinic acetylcholine receptor. It acts on antimuscarinic receptors reduces salivary, tracheobronchial, and pharyngeal secretions. Glycopyrrolate also decreases acid secretion in the stomach, thus preventing reflux acid secretions which can further reduce the irritation to the healing mucosa.

In our study we notice that there was significant reduction in the occurrence of PCF in post total laryngectomy with modified radical neck dissection patients when, intravenous sodium glycopyrrolate was given prophylactically. (Chi square test p value 0.0039). Also the odd ratio of 0.045 was suggestive that prophylactic use of sodium glycopyrrolate had a protective role and had odds of lesser chances of developing PCF.

To our knowledge, this is the first study aimed at the prevention of PCF in total laryngectomy cases. Our goal was to find effective method to prevent the PCF after total laryngectomy performed by residents of our institute, in whom previously we noticed high occurrence of PCF.

The reported higher rate of PCF occurrence could probably be attributed to poor nutritional status of majority of cases and also to the surgical expertise being less amongst the trainee residents. But with the prophylactic used of glycopyrrolate, even in the above scenario we noticed reduced occurrence of PCF after total laryngectomy. We recommend and encourage further studies on role of sodium glycopyrrolate in PCF prevention, with larger study setting and randomised prospective studies to increase the validity of our findings.

Conclusion

In our study of 50 patients who underwent total laryngectomy and modified radical neck dissection with adequate mucosal closure, we found injection glycopyrrolate significantly decreases the incidence of PCF. However this does not undermine the importance of meticulous mucosal closure with good suture material, broad spectrum antibiotics, anti-reflux prophylaxis, prevention of dead space under skin flap, appropriate preoperative nutrition, and control of infection in prevention of PCF. We have found no significant correlation between age, sex, staging, and site of tumor with incidence of PCF.

Compliance with ethical standards

Conflict of interest

All the authors declare they have no conflicts of interest and have not received any funding.

Ethical Approval

All procedures performed in the study were in accordance with ethical standards laid down in 1964 declaration of Helsinki and its later amendments or comparable ethical standards. The study was approved by institutional ethics committee. No animals were involved in the study.

Informed Consent

Written informed consent was obtained from all the individual participants in the study.