Hasan RA, LaRouere MJ, Kartush J, Bojrab D. Ambulatory Tympanomastoid Surgery in ChildrenFactors Affecting Hospital Admission. Arch Otolaryngol Head Neck Surg. 2004;130(10):1158-1162. doi:10.1001/archotol.130.10.1158
To identify clinical factors associated with postoperative nausea and vomiting (PONV) and failure to discharge from the hospital on the day of surgery in children undergoing tympanomastoid surgery.
Records of 144 children undergoing 152 tympanomastoid surgical procedures from July 1, 2001, through June 30, 2002, were retrospectively reviewed and the data analyzed.
A tertiary care university-affiliated hospital.
The mean ± SD age of the cases was 11 ± 3.7 years. Sixty-eight cases (45%) were middle ear procedures, while 84 cases (55%) were mastoid procedures. Forty-three cases (28%) were discharged home from the postanesthesia care unit (PACU), 55 cases (36%) were discharged on the day of surgery, and 142 cases (92%) were discharged home from the day surgery unit (DSU) within 23 hours after surgery. Patients who underwent mastoid procedures were more likely to require intravenous (IV) morphine sulfate in the PACU (75% vs 56%; P = .02) and were less likely to be discharged from PACU (15% vs 44%; P<.001) compared with patients who had middle ear procedures. In patients who underwent mastoid procedures, the presence of cholesteatoma (odds ratio, 1.9; 95% confidence interval, 1.0-3.7; P = .04) was associated with a higher likelihood of PONV. In both groups, the need for IV morphine sulfate to control pain on admission to DSU was associated with a higher occurrence of PONV. Factors that were significantly associated with failure to discharge from the hospital on the day of surgery were a pain score of 5 or greater, the presence of PONV, and the requirement of IV morphine sulfate on admission to DSU.
Factors associated with higher risks of PONV and failure to discharge from the hospital on the day of surgery include the presence of cholesteatoma, a pain score of 5 or greater, and the requirement of IV morphine sulfate at the time admission to the DSU.
Ambulatory otolaryngologic procedures continue to expand in the current health care system.1,2 Reduced cost and minimal disruption to the schedule of the child and the family are some of advantages of ambulatory surgery.1- 3
The type of surgical procedure is one variable that can significantly influence the nature of the postoperative course and the rate of postoperative complications.1- 4 Complex procedures, such as tympanomastoid surgery, may have a different postoperative course compared with other surgical procedures.5 Although life-threatening complications following ambulatory surgery in children are rare, minor problems and discomfort can occur.1- 4 For example, postoperative nausea and vomiting (PONV) is one of the most important reasons for unanticipated admission to the hospital in children undergoing ambulatory surgery.6,7 Studies on the factors associated with PONV and hospital length of stay (LOS) following tympanomastoid surgery in children are limited.5 The purpose of this study was to determine clinical factors that correlate with the occurrence of PONV and failure to discharge from the hospital on the day of surgery in children undergoing tympanomastoid procedures.
The institutional review board of Providence Hospital and Medical Centers approved the study. Consecutive patients with chronic otitis media, in whom medical therapy and ventilation tube insertion had failed, and who underwent tympanomastoid surgery between July 1, 2001, and June 30, 2002, were included in the study. Patients undergoing ventilation tube insertion or myringoplasty without other middle ear procedures were excluded.
Patients underwent a standard anesthesia procedure with tracheal intubation as follows. Anesthesia was induced with nitrous oxide and oxygen and increasing concentrations of sevoflurane. Anesthesia was maintained with sevoflurane, fentanyl citrate, oxygen, and air. Active drainage or the presence of granulation tissue in the middle ear cavity were the criteria for the use of perioperative antibiotics. The operative procedures consisted of middle ear procedures or mastoid procedures. Middle ear procedures included tympanoplasty with middle ear surgery, whereas mastoid procedures comprised tympanoplasty with atticotomy and/or antrostomy or mastoidectomy.
All procedures were done via a postauricular approach necessitating an incision behind the ear in addition to ear canal incisions. When necessary, ossicular chain reconstruction was performed in conjunction with tympanoplasty or tympanoplasty with mastoidectomy. The middle ear and ear canal were packed with an absorbable gelatin sponge (Gelfoam; Pfizer, New York, NY) prior to closure. Use of intraoperative antiemetics was left to the discretion of the otologist performing the surgery. After surgery, patients were admitted to the postanesthesia care unit (PACU), where they were observed for a minimum of 1 hour. They were subsequently transferred to a day surgery unit (DSU) if needed.
Postoperative pain was assessed based on the direct report method, using the Wong-Baker faces rating scale8 for children 8 years or younger, and the visual analogue scale for children older than 8 years.9,10 A pain score of 5 or greater was treated with intravenous (IV) morphine sulfate, ketorolac tromethamin, or a combination of both in that order. Patients who were able to tolerate oral pain medications received oxycodone hydrochloride. Sedation and anxiety scales (a scale of 1-4, where 1 = no symptoms and 4 = most severe) were recorded simultaneously along with the pain scale at least every 4 hours after surgery until the patient was discharged from the hospital. The presence of nausea and retching/vomiting and the appropriate medications used to ameliorate these symptoms were recorded. Medical records were reviewed and the following data were extracted: demographics, site and type of surgery, preanesthetic and anesthetic medications, length of anesthesia, length of surgery, intraoperative findings, intraoperative medications, and use of oxygen during the postanesthesia period.
The severity of middle ear disease was assessed by evaluating the presence of the following: frequent drainage (>4 times/year), tympanic membrane perforation, the presence of cholesteatoma in the middle ear cavity and mastoid air cells, facial nerve involvement by cholesteatoma, and granulomatous degeneration of the middle ear mucosa. Length of hospital stay, defined as the time between the end of anesthesia until the patient was discharged home, was documented and recorded.
Values are presented as mean ± SD. The χ2 and Fisher exact tests were used for nominal data and the unpaired t test was used for continuous data. Univariate and multivariate regression analyses were performed followed by stepwise logistic regression analysis to evaluate variables that were most closely associated with PONV and the need for overnight stay in the hospital after surgery. A statistical significance was assumed with P<.05.
The study population comprised 144 children undergoing 152 consecutive tympanomastoid surgical procedures. The mean age of the cases was 11 ± 3.7 years. The demographics and clinical characteristics are presented in Table 1. For the entire population of the study, length of anesthesia was 162 ± 55 minutes, the actual surgical time was 127 ± 54 minutes, and the hospital LOS was 15 ± 8 hours. Forty-three cases (28%) were discharged home from PACU, 55 cases (36%) were discharged on the day of surgery, and 142 cases (92%) were discharged home from the DSU within 23 hours after surgery. Of the 55 patients who were discharged home on the day of surgery, 40 underwent middle ear procedures, while 15 underwent mastoid procedures. The principal reasons for failure of discharge on the day of surgery were PONV (141 cases [92%]) and postoperative pain (124 cases [83%]). No difference in the occurrence of PONV was noted between boys and girls.
Patients who underwent mastoid procedures had a greater length of anesthesia and surgery, were more likely to receive intraoperative cefazoline sodium and IV morphine sulfate in the PACU, and were less likely to be discharged from PACU compared with patients who underwent middle ear procedures (Table 2). Dolasetron mesylate was used as a prophylactic antiemetic toward the end of surgery in 113 cases (74%) and did not have a statistically significant effect on the occurrence of PONV (P = .60). A higher proportion of patients who underwent mastoid procedures received dolasetron mesylate as a prophylactic antiemetic during surgery compared with patients who underwent middle ear procedures; however, there was no statistically significant difference (P = .30) between the 2 groups with regard to the occurrence of PONV that required antiemetics postoperatively (Table 2).
As given in Table 2, patients who underwent mastoid procedures required a higher number of doses of antiemetics to control PONV in the PACU. In this group, the presence of cholesteatoma (odds ratio, 1.9; 95% confidence interval, 1.0-3.7; P = .04) and the need for IV morphine sulfate for pain (odds ratio, 3.1; 95% confidence interval, 1.1-8.8; P = .03) at the time of admission to DSU were associated with a higher risk of PONV (Table 3). Also, in this group the presence of a pain score of 5 or greater at the time of admission to DSU, which was treated with IV morphine sulfate, was associated with failure to discharge from the hospital on the day of surgery (Table 4).
In children who underwent middle ear procedures, a pain score of 5 or greater and the need for IV morphine sulfate at the time of admission to DSU to treat such pain were associated with a higher risk of occurrence of PONV (Table 5). In this group, factors that were significantly associated with failure to discharge from the hospital on the day of surgery included a pain score of 5 or greater, the need for IV morphine sulfate, and the presence of PONV at the time of admission to DSU (Table 6).
There were no surgical complications such as bleeding, hematoma formation, or flap necrosis. In addition, there were no cases of iatrogenic inner ear injury such as unexplained sensorineural hearing loss or vertigo, and preoperative bone lines were preserved in all patients.
The occurrence of PONV and the hospital LOS was not different between canal wall up and canal wall down procedures. None of the children who were discharged home were readmitted to the hospital within 1 month after discharge.
Postoperative nausea and vomiting is one of the most common complications after ambulatory surgery, and it occurs following local, regional, and general anesthesia.6,7 Factors affecting PONV include age, sex, type of surgery, anxiety, history of motion sickness, postoperative pain, and use of opioids.6,7 The incidence of PONV is particularly common in patients undergoing otologic surgical procedures, for which an incidence as high as 80% has been reported.6,7 Megerian et al5 reported that a history of motion sickness was predictive of PONV and increased hospital LOS in adults undergoing tympanomastoidectomy.5 A history of motion sickness was not consistently available in this study because most children were either too young to understand the meaning of this syndrome or could not recall having experienced the symptoms of motion sickness.
The incidence of PONV is higher in children compared with adults, and in children the incidence increases with age to reach a peak in the preadolescent years.11 Most investigators have reported a significantly higher incidence of PONV in women than in men12; however, this sex difference has not been observed in the preadolescent age group.11,12 Our results are consistent with this observation.
In the present study, PONV was associated with the presence of significant pain that necessitated the use of IV morphine sulfate in the postoperative period. This is consistent with previous observations by Anderson and Krohg,13 who noted that relief of pain was associated with relief of nausea. The relationship between pain and the occurrence of PONV is further supported by the observation that the incidence of emesis in the postoperative period increases following reversal of opioid-mediated pain relief by naloxone.12
This study also demonstrates that children who undergo more complex middle ear surgical procedures, such as atticotomy, antrostomy, or mastoidectomy, are more likely to require overnight hospitalization following surgery. The need for more IV opioids for pain control and antiemetics for PONV are likely contributing factors.
In this study, there were no cases of iatrogenic inner ear injury leading to complications such as sensorineural hearing loss or vertigo. Preoperative bone lines were preserved in all patients. Therefore, PONV was most likely due to other factors such as perioperative medications and the stress induced on the ear by a combination of prolonged exposure, caloric and suction irrigation, and high-speed drilling of the bones surrounding the labyrinth.
Physical stimulation caused by drilling and irrigation of the bone adjacent to the inner ear has been implicated in the development of PONV after tympanomastoid surgery.14,15 Vibrations caused by drilling may influence the vestibular system.15 It has been shown that stimulation of inner ear structures by low-frequency sounds provoke postural instability in patients undergoing middle ear surgery for chronic otitis media.14,15 Suction and irrigation during surgery also stimulate the vestibular apparatus, which can further increase the incidence of PONV. Thus, it has been postulated that physical stimulation is one of the major causes of PONV after tympanomastoid surgery.14,16 This is supported by the fact that transdermal hyoscine, a drug used in motion sickness, is more potent in decreasing the incidence of PONV than ondansetron hydrochloride, which lacks antimotion sickness effects.17
At the request of the otologist performing the surgery, most children in this study received dolasetron mesylate (a 5-hydroxytryptamine type 3 [5-HT3] receptor antagonist) toward the end of the surgical procedure. However, there was no statistically significant difference (P = .60) in terms of the occurrence of PONV between patients who received dolasetron and patients who did not receive this antiemetic. Scuderi et al18 in a double blinded randomized and placebo-controlled trial demonstrated that there was little difference on the time to hospital discharge or return to normal activity when odansetron (also a 5-HT3 receptor antagonist) was administered at the beginning of the case and prior to induction of general anesthesia to a heterogeneous group of adults undergoing ambulatory surgery.
One of the limitations of the study by Scuderi et al18 was that the antiemetic was administered at the beginning of the case before induction of anesthesia. Sun et al19 showed that timing of the administration of a 5-HT3 receptor antagonist was relevant in relation to the development of PONV in patients undergoing otolaryngologic procedures. They demonstrated that ondansetron was more effective when administered at the end of surgery before the patient emerges from anesthesia.19 In our patients, dolasetron mesylate was administered toward the end of the surgical procedure, but it did not have a statistically significant impact (P = .60) on the occurrence of PONV. Megerian et al5 demonstrated that prophylactic administration of antiemetics did not positively influence the incidence of PONV.5 Patients in this study were primarily adult patients, but the study also included a small number of children undergoing tympanomastoid surgery.5 Megerian et al5 did not use dolasetron or other 5-HT3 receptor antagonists; however, they did use other antiemetics, primarily droperidol, in 45% of their patients.
We do not dispute the distressing effects of PONV and we believe that elimination of these symptoms is still a worthwhile endeavor. However, based on our study and previous studies,5,18 it does not appear that prophylactic administration of an antiemetic would eliminate PONV or result in better outcome such as time to discharge from the hospital.
A significant proportion of children were discharged from the hospital on the day of surgery. None of these children had to be readmitted to the hospital, an observation that suggests that selective discharge to home on the same day of surgery is possible in children undergoing tympanomastoid surgery. To our knowledge, this is one of the first studies to examine the potential for same-day discharge for children undergoing tympanomastoid surgery. Megerian et al5 reported that up to 33% of patients undergoing tympanomastoidectomy were discharged from the PACU. The average age of patients in that study was 34.5 years, with a range of 2 to 73 years, suggesting that there were some pediatric patients as well.
We sought to determine whether the severity of middle ear disease, the length of anesthesia, and the surgical procedure could affect PONV and the hospital LOS. This information is potentially helpful in preoperative counseling of families and obtaining approval from third-party payers for hospital stay after surgery. The presence of cholesteatoma was associated with a higher risk of occurrence of PONV in patients undergoing mastoid procedures. This may be related to the need for more manipulation of and instrumentation of middle ear cavity and its structures.
Because of the retrospective nature of this study, other important influences on hospital utilization, such as comorbid medical conditions and family support, were not assessed. Patients with low or no comorbidity and excellent family support are more likely to leave the hospital earlier. Other limitations of this study include the fact that it is from a single institution and the pattern of practice at our institution may be different from that at other institutions.
Tympanomastoid surgery in children can safely be performed on an outpatient basis. Postoperative pain requiring IV opioids and PONV are 2 principal reasons for failure to discharge from the hospital on the day of the surgery. Most children (92%) can be discharged home from the hospital within 23 hours after surgery. Children who undergo more complex middle ear surgical procedures are more likely to require an overnight observation. Children with an underlying cholesteatoma are at a higher risk of PONV. Prophylaxis with a 5-HT3 receptor antagonist had no effect on the occurrence of PONV or length of hospital admission.
Correspondence: Rashed A. Hasan, MD, Providence Hospital and Medical Centers, 16001 W Nine Mile Rd, Southfield, MI 48075 (email@example.com).
Submitted for publication September 10, 2003; final revision received March 19, 2004; accepted April 9, 2004.