Early post-operative morbidity following tonsillectomy in children: implications for day surgery

Adenotonsillectomy and tonsillectomy are amongst the most commonly performed surgical procedures. In our prospective study we looked at early post-operative morbidity (haemorrhage, vomiting and pyrexia) in 521 consecutive children following tonsillectomy, adenotonsillectomy and adenoidectomy. While the incidence of reactionary haemorrhage is low, pyrexia and vomiting are common. The implications of this high early post-operative morbidity for day surgery are discussed.     

Adult day case tonsillectomy: a safe and viable option

Day case tonsillectomy, in general, is not considered a routine procedure. Many otolaryngologists would not perform this operation on an out-patient basis due to the risk of reactionary haemorrhage after discharge from hospital. A prospective trial of carefully selected adult patients undergoing day case tonsillectomy was performed. Each was given a questionnaire to assess post-operative morbidity including pain, time required to return to normal diet and comments on the ambulatory aspect of their surgery. The results of 48 patients are analysed. None of these had haemorrhage within 24 h of their operation. Twenty patients (42%) were happy to have the procedure performed in this way whilst 19 (39%) found it tolerable. Two patients had to be admitted overnight as they had not recovered enough to be discharged. It is concluded that adult day case tonsillectomy in selected patients is a safe and viable option.     

Post tonsillectomy haemorrhage in children: The value of routine monitoring of the pulse

Twenty-one children with reactionary haemorrhage after tonsillectomy were returned to theatre for the control of this haemorrhage. This group was matched for age and weight with 21 patients who underwent an uncomplicated tonsillectomy. The maximum pulse rate, the mean pulse rate and the maximum rise in pulse rate were recorded during the post-operative period and were noted for each group. No statistical difference existed between the groups. The monitoring of children in the period following tonsillectomy may be inadequate if the pulse alone is measured.     

Bipolar diathermy or ligation for haemostasis in tonsillectomy? A prospective study on post-operative pain.

Forty-nine children for tonsillectomy were entered in a prospective controlled single blind study to compare the post-tonsillectomy morbidity of bipolar diathermy as opposed to ligation for haemostasis. We found no significant difference in post-operative discomfort nor reactionary or secondary haemorrhage in the two methods used. This is the first controlled trial where bipolar, rather than unipolar diathermy is used, and compared with ligation in the operation of tonsillectomy.     

Salicylates and post-tonsillectomy haemorrhage

A prospective study of 712 patients who had tonsillectomy by guillotine or dissection suggests that salicylates used as post-tonsillectomy analgesia increase the post-operative haemorrhage rate. Paracetamol used as post-operative analgesia was shown to be related to a significantly reduced haemorrhage rate. Guillotine tonsillectomy was not shown to have any adverse effect on the haemorrhage rate.     

Classification of haemorrhage following tonsillectomy.

To date there exists no sensible way of classifying the intensity of haemorrhage following tonsillectomy, though this is a prerequisite when comparing the results presented in literature. We evaluated the incidence of post-operative haemorrhage according to our classification in 602 patients who underwent tonsillectomy at our department in 1999. In 21 patients a grade 1 (spontaneous cessation) bleeding occurred. One patient had a grade 2 (infiltration anaesthesia), 14 patients had a grade 3 bleeding (treatment under general anaesthesia) with one patient receiving a blood transfusion. Two patients had a grade 4 bleeding (ligature of the external carotid artery). There was no patient with grade 5 bleeding (lethal outcome). Grades 1 and 3 had the same incidence rates. Primary haemorrhage (<24 hours) can be expected in the majority of patients undergoing tonsillectomy. Secondary haemorrhage is rare and can be life-threatening.     

A prospective study of factors which may predispose to post-operative tonsillar fossa haemorrhage.

Adeno-tonsillectomy and tonsillectomy are frequently performed and haemorrhage remains an important post-operative complication. A prospective study of 1090 patients undergoing surgery over a 12-month period was undertaken investigating whether the following factors influenced post-operative haemorrhage: age and sex of patients, method of tonsillectomy, haemostasis and premedication, month of operation and experience of surgeon. A significantly higher incidence of post-operative haemorrhage occurred in the group aged 16-25 years and significantly more males suffered post-operative haemorrhage. This haemorrhage more frequently occurred in the warmer months of May to July. Guillotine tonsillectomy was found to be as effective as dissection and no higher incidence of post-operative haemorrhage was recorded. Method of haemostasis, surgical experience and premedication did not influence the outcome. We conclude that the guillotine used in controlled circumstances is as safe as dissection.     

A prospective study of factors which may predispose to post-operative tonsillar fossa haemorrhage

Adeno-tonsillectomy and tonsillectomy are frequently performed and haemorrhage remains an important post-operative complication. A prospective study of 1090 patients undergoing surgery over a 12-month period was undertaken investigating whether the following factors influenced postoperative haemorrhage: age and sex of patients, method of tonsillectomy, haemostasis and premedication, month of operation and experience of surgeon. A significantly higher incidence of post-operative haemorrhage occurred in the group aged 16–25 years and significantly more males suffered post-operative haemorrhage. This haemorrhage more frequently occurred in the warmer months of May to July. Guillotine tonsillectomy was found to be as effective as dissection and no higher incidence of post-operative haemorrhage was recorded. Method of haemostasis, surgical experience and premedication did not influence the outcome. We conclude that the guillotine used in controlled circumstances is as safe as dissection.     

Post tonsillectomy morbidity: following the introduction of single-use-instruments

OBJECTIVE: Following the introduction of single-use-instruments (SUI) for tonsillectomy in January 2001 in UK, the Medical Devices Agency (MDA) received reports of significant increase in secondary haemorrhage in some centres. This was believed to be related to electrosurgical (diathermy) forceps. The aim of this study is to compare post tonsillectomy morbidity before (year 2000) and after (year 2001) the introduction of SUI in a unit where bipolar dissection is the most popular method of dissection. METHOD: The case notes of all tonsillectomy patients between 20 April and 31 October for the years 2000 (145 patients) and 2001 (166 patients) were reviewed. The patients' age and sex, other concurrent operations, grade of surgeon, dissection and haemostasis methods, length of post operative hospital stay and the reason for prolonged stay (if more than one night) were noted. Readmitted cases were analysed to identify the cause and the management. RESULTS: There was no case of reactionary haemorrhage in both the non SUI (year 2000) and the SUI (year 2001) groups. For the non SUI versus SUI group, 6.2 versus 4.8% stayed more than one night post operatively; 4.8 versus 5.4% were readmitted for secondary haemorrhage with 1.4 versus 2.4% return to theatre to control bleeding, respectively. The average hospital stay for the readmitted cases were 2.1 versus 1.6 days and the average time lapse between tonsillectomy and secondary haemorrhage were 6.7 versus 6 days respectively. CONCLUSION: There was no significant difference in post tonsillectomy morbidity following the introduction of SUI.     

Increased post-operative haemorrhage seen in adult coblation tonsillectomy

Coblation is a new soft tissue surgical technique that is being used for tonsillectomy. Published results show a significant decrease in the amount of post-operative pain experienced by patients undergoing coblation tonsillectomy. There has been no published work to date on the incidence of post-operative haemorrhage. From August 2001 to November 2002 one surgeon performed 36 coblation tonsillectomies on adults. On another list he performed 29 by his standard method of dissection and bipolar coagulation. Retrospective analysis found a significant increase in the secondary haemorrhage rate in adult patients undergoing coblation tonsillectomy (22.2 vs. 3.4 per cent). At our department coblation tonsillectomy has been abandoned until further work into its safety has been published.     

Does synchronous nasal surgery increase the risk of post-operative haemorrhage in adult patients undergoing tonsillectomy?

Otolaryngologists occasionally perform simultaneous procedures on their patients, the indication being co-existent pathology. This is not a universally accepted practice as some feel that the post-operative morbidity is increased in such patients. Tonsillectomy is a common otolaryngological procedure and the main life-threatening complication, haemorrhage is easily identifiable. This study compares the incidence of post-operative haemorrhage requiring admission in patients undergoing tonsillectomy alone with that in patients undergoing tonsillectomy and nasal surgery. A retrospective study was undertaken of all adult patients undergoing tonsillectomy alone or tonsillectomy and synchronous nasal surgery in Aberdeen Royal Infirmary. Seventy-one patients (34 male) with a mean age of 23 years underwent tonsillectomy and synchronous nasal surgery over a nine-year period 1991-1999. The commonest nasal procedure was reduction of the inferior turbinates (in 48 cases). Three hundred and ninety-eight patients (131 male) with a mean age of 23 years underwent tonsillectomy alone over a nine-month period, July 1998 to April 1999. The incidence of post-tonsillectomy haemorrhage in the synchronous nasal surgery group was 12.7 per cent (9 out of 71) compared with four per cent (16 out of 398) in the tonsillectomy alone group (p <0.01, Chi-squared test). There was no difference in hospital stay between the two groups (mean two days). Synchronous nasal surgery increases the incidence of post-tonsillectomy haemorrhage.     

Coblation tonsillectomy: is it inherently bloody?

The aim of the study was to compare a single surgeon’s post-tonsillectomy haemorrhage rates using cold steel dissection and coblation tonsillectomy techniques. Retrospective study on patients, who underwent tonsillectomy at West Wales General Hospital (WWGH) performed by a single surgeon from 2006 to 2010 employing both cold steel and coblation tonsillectomies. Data were analysed using Mann–Whitney and Chi-squared tests. The nominated surgeon performed 239 tonsillectomies at WWGH from 2006 to 2010. 119 patients underwent cold steel dissection and 120 had coblation tonsillectomy. There was no demographic difference between the two groups. There was no statistically significant difference in the length of hospital stay between the two groups (median 1 day in each group). 6/119 (5.0%) patients in the cold steel group, and 7/120 (5.8%) in the coblation group had post-operative bleeding (p = 1.00). The return to theatre rate for cold steel dissection was 1/119 (0.84%) and for coblation surgery was 1/120 (0.83%) (p = 1.00). Among the first 60 cases of coblation tonsillectomies, 4 patients (6.6%) had post-operative haemorrhage and the latter 60 cases had 3 patients (5%). There was no evidence of a difference in the overall post-operative bleeding between those who had cold steel dissection and coblation tonsillectomies. These data suggest that higher post-operative haemorrhage is not inherent to coblation tonsillectomy.     

Is the KTP laser effective in tonsillectomy?

Thirty-eight patients underwent a randomized double-blind trial using the KTP laser for tonsillectomy on one tonsil and standard dissection tonsillectomy on the other tonsil. Blood loss was less on the laser side. However, pain though initially slightly less on the laser side (days 1 and 2 post-operation) was worse on the laser side at 2 weeks due to delayed healing of the tonsillar bed. There were no primary or reactionary haemorrhages but a 15% incidence of secondary haemorrhage on the laser side.     

Post-tonsillectomy haemorrhage. A retrospective study of 1150 operations

The medical records of 1150 tonsillectomized patients were retrospectively reviewed for registration of the frequency of post-tonsillectomy haemorrhage in relation to sex and age of the patients, the indications for tonsillectomy, and the operative experience of the surgeon. The number of post-operative bleeds requiring surgery was 32 (2.8%) and occurred most frequently in young men and in patients with a history of previous peritonsillar abscess undergoing cold tonsillectomy. The frequency of haemorrhage in abscess tonsillectomy was not higher than expected. As previous studies have demonstrated that abscess tonsillectomy is associated with a minimal risk of spreading the infection or other serious complications, it is suggested that abscess tonsillectomy should be the preferable treatment of peritonsillar abscess.     

A prospective double‐blind randomized controlled trial comparing the suitability of KTP laser tonsillectomy with conventional dissection tonsillectomy for day case surgery

Tonsillectomy using a KTP laser has been performed increasingly but is not a routinely practised technique in the UK. In the USA, tonsillectomy is often performed as a day case procedure but, here in the UK, it is still standard practice to admit patients for overnight stay. We present the largest prospective double‐blind randomized controlled trial to date (151 patients) comparing KTP laser with standard dissection tonsillectomy and assess the suitability of both procedures for day case surgery. We found that there was significantly less peroperative haemorrhage if tonsillectomy was performed using the KTP laser, but it did cause more postoperative pain, more depression in mood and a higher rate of both reactionary and secondary haemorrhage, which was not significant when compared with conventional dissection. There was no difference in operating time, and over 40% of patients in each group needed overnight admission. We conclude that KTP laser tonsillectomy offers no benefit apart from less intraoperative bleeding over standard dissection tonsillectomy. Discharge from hospital after tonsillectomy was found to be unpredictable. Tonsillectomy is therefore an unsuitable procedure for planned surgery through a day unit, but approximately 58% of patients could be discharged on the same day from an extended day surgery unit, and the rest have one night in hospital.     

A prospective double-blind randomized controlled trial comparing the suitability of KTP laser tonsillectomy with conventional dissection tonsillectomy for day case surgery

Tonsillectomy using a KTP laser has been performed increasingly but is not a routinely practised technique in the UK. In the USA, tonsillectomy is often performed as a day case procedure but, here in the UK, it is still standard practice to admit patients for overnight stay. We present the largest prospective double-blind randomized controlled trial to date (151 patients) comparing KTP laser with standard dissection tonsillectomy and assess the suitability of both procedures for day case surgery. We found that there was significantly less peroperative haemorrhage if tonsillectomy was performed using the KTP laser, but it did cause more postoperative pain, more depression in mood and a higher rate of both reactionary and secondary haemorrhage, which was not significant when compared with conventional dissection. There was no difference in operating time, and over 40% of patients in each group needed overnight admission. We conclude that KTP laser tonsillectomy offers no benefit apart from less intraoperative bleeding over standard dissection tonsillectomy. Discharge from hospital after tonsillectomy was found to be unpredictable. Tonsillectomy is therefore an unsuitable procedure for planned surgery through a day unit, but approximately 58% of patients could be discharged on the same day from an extended day surgery unit, and the rest have one night in hospital.     

Bipolar scissors versus cold dissection for pediatric tonsillectomy--a prospective,randomized pilot study

OBJECTIVE: To evaluate bipolar scissors tonsillectomy by comparing it with traditional cold dissection tonsillectomy. The outcome measures used were: (1) intra-operative bleeding; (2) operative time; (3) post-operative pain; and (4) complication rates including reactionary and secondary hemorrhage. METHOD: A prospective, randomized multiunit study involving three teaching hospitals in Belfast. Fifty consecutive children aged 10-16 years undergoing tonsillectomy for recurrent or chronic tonsillitis, between March 2000 and September 2000 were recruited as a subgroup of 200 patients selected for this study. These children were analysed separately from the adults, in a pilot study for the above parameters. RESULTS: The mean age of the study population was 14.3 years. Sixty-eight percent of the children were girls. Median intra-operative blood loss was 6 ml for bipolar scissors tonsillectomy and 86 ml for cold dissection tonsillectomy (P<0.001). The median operative time was 10.5 min for bipolar scissors tonsillectomy compared to 14.5 min for the cold dissection method (P=0.001). There was no statistically significant difference in the pain scores between the two methods (P>0.05). The overall reactionary hemorrhage rate was 4% while the overall secondary hemorrhage rate was 14%. The hospital readmission rate was 4%. The reactionary and secondary hemorrhage rates were unaffected by the surgical method. CONCLUSIONS: This pilot study has shown that bipolar scissors tonsillectomy is a relatively safe technique in children aged 10-16 years with a similar morbidity to the cold dissection method. Its use is associated with a significant decrease in surgical time and blood loss compared to the cold dissection method. These advantages make it a favourable instrument for pediatric tonsillectomy especially in this age group.     

Return to theatre in secondary post-tonsillectomy haemorrhage: a comparison of coblation and dissection techniques

The aim of this study is to examine the incidence of return to theatre (RTT) for post-operative haemorrhage following coblation and dissection tonsillectomy and to investigate those that required RTT more than 10 days post-surgery. Retrospective review of post-tonsillectomy haemorrhages requiring RTT from April 2005 to March 2009 was conducted. Of 2,541 tonsillectomies performed, 81% were by coblation and 19% by dissection methods. The overall RTT rate was 1.7%. No difference was found in the overall RTT rates for primary and secondary haemorrhage between the two techniques. However, the overall RTT rates for primary and secondary haemorrhage were higher in adults than children (P = 0.0456 and P = 0.0215, respectively). RTT for secondary haemorrhage during the first ten post-operative days occurred in both coblation and dissection tonsillectomy with no significant difference. After the first post-operative week, late secondary bleeding requiring RTT occurred only in the coblation group (P = 0.0676). Four patients required blood transfusion; all were in the coblation group, three of which were required during RTT in the late secondary haemorrhage (after 10 days). The post-operative RTT rates for coblation tonsillectomy did not reveal a change of trend over the 4-year study period. Our RTT rate for secondary haemorrhage is higher than earlier published results. A learning curve could not be identified in RTT for coblation tonsillectomy haemorrhage. Late secondary haemorrhages requiring surgical intervention have only been identified in cases performed by coblation and could potentially be life threatening as 33% (3/9) required blood transfusion. This phenomenon may be explained by a particular physiological healing process associated with coblation.     

Day-case tonsillectomy-is it appropriate?

There is continued encouragement to increase the use of day surgery. Recent publications have suggested that day-case tonsillectomy is a safe procedure due to the low primary haemorrhage rates (0.14–3.5%). One of the suggested benefits of day surgery is that patients want it. They prefer to recover at home after an operation. With tonsillectomy, personal experience suggested that this was not the case. A review of 117 patients having tonsillectomy was undertaken. All patients stayed in for at least one post-operative night. No patients or parents thought that the post-operative stay was too long (80%‘just right', 20%‘too short') and only 7% would have been happy to go home on the day of operation. ‘Safety’ does not automatically make an operation suitable for day-case surgery. Pain, nausea, vomiting, drowsiness and anxiety about the operation and post-operative course were all reasons given for not wanting to go home on the day of surgery. The justification for the increased use of day surgery is that it increases efficiency by reducing costs per case while maintaining the quality of care. One aspect of quality of care is patient acceptability and before day-case tonsillectomy is acceptable to patients the factors responsible for the post-operative morbidity need to be addressed.     

Analgesia in tonsillectomy: a double-blind study on pre and post-operative treatment with diclofenac.

The efficacy and tolerability of diclofenac suppositories given pre and/or post-operatively were investigated in a randomized double-blind study on 99 patients undergoing tonsillectomy. In one group, 50 mg diclofenac was given 1 h preoperatively, followed by 50 mg directly after the operation. In another group, diclofenac 100 mg was given only immediately post-operatively. A significantly lower consumption of rescue analgesics (paracetamol and/or pethidine) was found in the group treated preoperatively with diclofenac and the average time until first demand of rescue analgesics was significantly longer compared to the group given diclofenac post-operatively only. The tolerance was good and no serious bleeding complications occurred in either group. In the treatment of post-operative pain after tonsillectomy, the combination of pre and post-operative administration of diclofenac suppositories resulted in significantly lower consumption of rescue analgesics and is thus preferable to administration solely post-operatively.