The Role of Second-Look Nasal Endoscopy After Pediatric Functional Endoscopic Sinus Surgery

Objective  To determine if second-look nasal endoscopy (SLE) improves clinical outcome after pediatric functional endoscopic sinus surgery (FESS).

Design  A retrospective review was performed for all patients who had undergone FESS between January 1993 and December 1994. The patients were divided into 2 groups: those who underwent SLE 2 to 3 weeks after FESS (n=94) and those who did not undergo SLE after FESS (n=53). Patients with a history of cystic fibrosis, bone marrow transplantation, or isolated sphenoidal sinus disease were excluded from the study.

Setting  Pediatric tertiary care hospital.

Patients  All patients had a diagnosis of chronic rhinosinusitis, had failed medical therapy, and had subsequently undergone FESS.

Main Outcome Measures  The need for revision sinus surgery was used as the definition for poor clinical outcome, indicating that sinus disease recurred or was persistent. The number of patients requiring revision sinus surgery in each group was compared.

Results  Revision surgery was performed in 20 of the 94 patients who underwent SLE after FESS. Revision surgery was performed in 10 of the 53 patients who did not undergo SLE after FESS. Six patients in the "no-second-look group" underwent follow-up nasal endoscopy at some time after the initial FESS. There was no significant difference in terms of clinical outcome between the 2 groups.

Conclusions  The rate of revision sinus surgery was comparable in those patients who underwent SLE after FESS and those who did not. The usefulness and application of SLE need to be considered carefully.

PEDIATRIC functional endoscopic sinus surgery (FESS) is now commonly performed when prolonged medical therapy to treat rhinosinusitis is ineffective. The procedure was first described in 1989 in a series of 57 patients in whom the procedure was performed safely.¹ A description of the technique and postoperative care included nasal endoscopy performed 2 weeks after the initial procedure with the patient under general anesthesia. This second procedure, or "second look," consisted of the removal of any synechia, clots, or granulation tissue. Other investigators have performed this procedure to remove the stents placed at the time of initial FESS and to clean the ethmoid cavities and antral ostia of any gross granulation tissue.² The idea of second-look nasal endoscopy (SLE) was to allow postoperative nasal cleaning, which, theoretically, would improve surgical outcome.

Numerous publications describe the routine use of SLE.^3,4 To our knowledge, there have been no studies examining the need for SLE. The purpose of this study was to determine whether SLE improves clinical outcome after pediatric FESS.³

We performed a retrospective chart review of all cases involving pediatric patients who underwent FESS at Children's Hospital Medical Center, Cincinnati, Ohio, between January 1,1993, and December 31,1994. Children with cystic fibrosis or isolated sphenoidal sinus disease were excluded. The patients were divided into 2 groups: those who underwent SLE 2 to 3 weeks after FESS and those who did not (NoSLE). Four attending pediatric otolaryngologists with the same referral patterns operated on the patients. Three of the 4 staff members performed routine second-look procedures, and 1 did not. Information was gathered from the patient records regarding the patient's age at the time of the initial FESS; any history of allergy, asthma, or immune deficiency; severity of disease based on pre–initial FESS computed tomographic (CT) scan; and the type of initial FESS that was performed. The identification of patients with allergy, asthma, or immune deficiency was based on a complete history from the parents as well as on a review of the pediatricians' records on referral to the otolaryngologist. The CT scan findings were divided into 4 groups: (1) no disease, (2) mild (1 sinus cavity per side) disease, (3) moderate (more than 1 sinus cavity per side) disease, or (4) severe (pansinus) disease. Information was recorded as to the findings on SLE, if the procedure was done. Records were studied to identify which patients required revision sinus surgery (RS). Nasal endoscopy alone performed 4 weeks or more after FESS was not considered SLE or RS and was recorded separately.

All patients met the criteria to become candidates for initial FESS. The majority of patients were referred by their primary care provider with chronic sinus complaints that were present for at least 3 months. All patients had been treated with a β-lactamase–stable antibiotic for a minimum of 6 weeks. At the completion of the antibiotic course, nearly all patients had evidence of mucosal thickening (>4 mm) on CT scans of the sinuses, as well as persisting signs and symptoms of chronic rhinosinusitis. Topical nasal steroid therapy was used in the majority of patients for at least 3 weeks before they became candidates for FESS.

To become a candidate for RS, the patient must have had documentation of recurrent signs and symptoms of chronic rhinosinusitis that failed to respond to medical treatment (with a β-lactamase–stable antibiotic) for more than 12 weeks. Topical nasal steroids were used in the majority of patients for at least 3 weeks before they became candidates for RS. Pediatric FESS and RS were performed on all of these patients by attending staff or an upper-level physician in training under the direct supervision of the attending staff. All surgeons follow the same technique for pediatric FESS as described by Lusk.⁴

Statistical analysis was performed using frequency tabulations with the χ² test, Kendall τ nonparametric correlation coefficients, and stepwise logistic regression analysis to determine if any significant differences were present between the 2 groups (SLE and NoSLE) in regard to all categories of data gathered.

A total of 147 patients were identified and reviewed. No complications, such as cerebrospinal fluid leak or orbital injury, were encountered at the time of initial FESS or RS. The patients were divided into 2 groups for analysis and comparison: the SLE group and the NoSLE group. The average length of follow-up from the initial FESS to the last documented clinic visit was 19.0 months (range, 6-48 months). For the great majority of the patients, follow-up continued via the primary care physician for a minimum of 2 years.

Of the 147 patients, 98 were male and 49 were female. The age range at the time of initial FESS was 1 to 20 years (average age, 5.69 years). The most common presenting signs and symptoms for all patients in the study prior to the initial FESS were purulent nasal discharge, nasal congestion, cough, and recurrent ear infections.

Eighty-two patients (55.8%) had allergies, 28 (19.0%) were asthmatic, and 13 (8.8%) had immune deficiencies. The presence of allergies was found to correlate statistically (0.188) with the presence of asthma (P<.05). Each of these associated medical conditions was equally distributed between the SLE group and the NoSLE group.

Three (2.0%) of the 147 patients had no disease on CT scan prior to the initial FESS, while 17 (11.6%) had mild disease, 89 (60.5%) had moderate disease, and 38 (25.9%) had severe disease. The 3 patients with no findings on CT scan all had classic signs and symptoms of chronic rhinosinusitis and thus underwent FESS. The severity of disease based on CT scan was equally distributed between the SLE group and the NoSLE group.

The type of initial FESS procedures included unilateral or bilateral anterior or anterior and posterior ethmoidectomy with maxillary sinus antrostomy.

Statistical analysis showed that the procedures were equally distributed between the SLE group (n=94) and the NoSLE group (n=53). The type of procedure performed at initial FESS was not found to be significantly associated with RS. The findings on SLE included clean cavity, synechiae, granulation tissue, inflammatory or polypoid mucosa, mucoid or mucopurulent drainage, and scarring (Table 1).

Overall, 30 of the 147 patients required RS (not including nasal endoscopy): 20 (21.3%) in the SLE group and 10 (18.9%) in the NoSLE group. An additional 3 patients in the SLE group and 6 patients in the NoSLE group did undergo nasal endoscopy at least 4 weeks after the initial FESS, but findings at that time did not indicate that RS was necessary. Statistical analysis of these data revealed no association between NoSLE and RS.

The average time between initial FESS and RS for all patients was 14.2 months. In the SLE group, the average time between initial FESS and RS was 15.7 months, and in the NoSLE group, it was 13.8 months. Revision procedures included lysis of synechiae, unilateral or bilateral anterior or anterior and posterior ethmoidectomy with maxillary sinus antrostomy, antral washout, maxillary antrostomy, and nasal endoscopy (Table 2). Antral washouts were performed only in the SLE group, owing to the practice philosophy of one of the surgeons. Endoscopic findings at RS included synechiae, granulation tissue, inflammatory or polypoid mucosa, mucoid or mucopurulent drainage, scarring, and mucocele formation (Table 3).

Additional statistically significant findings included a positive correlation between RS and severity on CT scan (0.219, with P=.006). A negative correlation was identified between RS and age at the time of FESS (−0.147, with P=.04), as well as between severity on CT scan and age at the time of FESS (−0.229, with P=.001). In other words, RS was more common and CT scan findings were more severe in younger children.

After compiling the data listed, we went back to compare the number of general anesthetics administered to the SLE group and the NoSLE group. In the SLE group, 211 anesthetics were administered to 94 children, and in the NoSLE group, 69 anesthetics were administered to 53 children (anesthetic-child ratio, 2.24 in the SLE group and 1.31 in the NoSLE group).

Pediatric FESS is generally accepted and supported as a treatment for refractory pediatric rhinosinusitis.⁵ The definition of what constitutes adequate or appropriate medical therapy prior to pediatric FESS remains controversial. Since the inception of pediatric FESS, SLE has become common practice. The rationale appears to stem from adult literature in reference to the importance of aggressive nasal hygiene in the postoperative period. Stammberger⁶ states that follow-up treatment of adult patients after FESS, consisting of cleaning crusted blood and eschar, is essential. He begins this treatment 2 to 4 days after surgery and continues it every 3 to 5 days for the next 10 days. Other investigators have emphasized that the majority of postoperative cavities require little care.⁷ A recent study from the United Kingdom concluded that minimal postoperative follow-up after adult FESS allowed results comparable with those of other adult series that used aggressive postoperative follow-up.⁸

To our knowledge, there have been no prospective studies performed to ascertain the necessity and/or benefit of SLE. All current reports on technique and postoperative recommendations after pediatric FESS, however, state that SLE should be performed 2 to 3 weeks after surgery for toilet, debridement, and removal of middle meatal splints. Even when absorbable stenting material is used, it is stated that children greatly benefit from this procedure.⁹ Findings at SLE usually include "some" granulation tissue and early adhesions between the middle turbinate and the lateral nasal wall.¹⁰ Stankiewicz¹¹ reported SLE findings in 34 pediatric patients, including a well-healed or patent ostia on 32 sides; granulation tissue, synechiae, or scarring on 27 sides; and evidence of the antrostomy "closing" on 30 sides. He commented that, because no study has identified objective findings after pediatric FESS, scarring, synechia, or antrostomy closure may be present in many cases after FESS or SLE and that children possibly tolerate these conditions without clinical or symptomatic relevance.

The authors of a report on FESS in developmentally delayed children opted not to perform SLE in 5 of 7 patients because of anesthetic risks.¹² Follow-up on these 7 patients found recurrent sinus disease only in the 2 patients who did undergo SLE.

The use of SLE in the pediatric population requires general anesthesia and is usually performed with endotracheal tube intubation to prevent the aspiration of blood. The need for a second general anesthetic raises the issue of the relative risks and benefits of SLE. Although pediatric anesthesia is relatively safe, morbidity and mortality are still associated with it. The incidence of major anesthesia-related complications, defined as any fatal or life-threatening event, has been reported at a rate of 7 per 10000 anesthetics administered.¹³ In addition, an estimated cost of nasal endoscopy would include a surgeon's fee of $400, an anesthesia fee of $320, and an operating room fee of $280 per patient.Indirect costs, such as parental time off work or school and transportation expenses, also have to be taken into consideration.

It has been our experience that, by 2 weeks after FESS, the sinus cavity is usually in a healing phase, denuded, with granulation tissue being present. Our observations are supported by the fact that reepithelialization will take at least 2 to 3 weeks after surgery to occur, and may take as long as 6 weeks.⁷ Most often, instrumentation of the middle meatus region alone causes bleeding and poor visualization. Furthermore, when SLE is completed, a cavity with freshly denuded mucosa remains, which will likely form more granulation tissue, take another 2 to 3 weeks to reepithelialize, and possibly form new synechiae.

To our knowledge, this is the first study to compare a group of patients who underwent SLE with a group who did not. Our results showed no significant difference in the rate of RS between the 2 groups. Furthermore, the number of general anesthetics per child was far greater in the SLE group (2.24) than in the NoSLE group (1.31).Our findings at the time of RS demonstrated a higher rate of synechia formation in the group that did not undergo SLE, but within the time frame of our follow-up, this did not appear to affect the rate of RS. Longer-term follow-up would be necessary to definitively address this issue.

Based on a review of the literature, the estimated success rate for FESS in children is 80%.^14-16 The range of pediatric patients requiring revision FESS is 8% to 32%.^10,17,18 After revision FESS, the failure rate, based mainly on adult studies, may range from 22% to 30%.^19,20 Based on an outcome study of rhinosinusitis in children, although there is an 85% to 100% chance of symptoms improving after FESS, there is only a 15% to 60% chance of cure.⁵

In our series, 20.4% of all patients underwent RS. The majority had "success" after the initial FESS. However, a number of children, although improved, continued to have signs and symptoms of chronic rhinosinusitis that did not respond to medical therapy and thus fit our criteria for RS. Revision FESS is indicated if symptoms of chronic rhinosinusitis persist after the primary FESS and optimal medical treatment.¹⁹ Although the rate of revision FESS procedures in children is not the topic of this article, it is an issue that needs to be addressed and studied carefully. Our patient population included a high number of children with allergies (55.8%), which implies an underlying systemic condition that may predispose to persisting rhinosinusitis-related complaints following FESS.

There are certain circumstances in which SLE is appropriate and even necessary. These include cases that do not allow adequate completion of the initial FESS owing to excessive bleeding, extensive disease, or poor visualization, or for anesthetic reasons. If an excessive area of raw surface is noted and long-term stenting with gel film or other material is deemed appropriate for that particular patient, SLE may be warranted. We do recommend aggressive postoperative use of hypertonic saline solution.

Second-look nasal endoscopy is not necessary in all children. Administration of a second anesthetic can be avoided in at least 50% of patients, with no sequelae. Second-look surgery or RS should be reserved for symptomatic individuals and performed at a time that is appropriate for each individual patient based on his or her clinical state. The 3 pediatric otolaryngologists from our institution who previously performed routine SLE now follow the recommendations listed above based on the findings of this study.

In the majority of FESS cases, in which careful, meticulous, and complete dissection has been performed, we believe that SLE is not necessary. A small number of patients may develop recurrent sinus symptoms and require follow-up nasal endoscopy. However, it is our opinion that these cases should be dealt with on an individual basis. Although not ideal or definitive, our retrospective study provides insight into the concept of SLE. The application and usefulness of this technique should be considered carefully.

Accepted for publication October 3, 1997.

Presented in part at the annual meeting of the American Society of Pediatric Otolaryngology, Scottsdale, Ariz, May 14, 1997.

Corresponding author: David L. Walner, MD, Department of Otolaryngology–Bronchoesophagology, Rush-Presbyterian-St Lukes Medical Center, 1653 W Congress Pkwy, Chicago, IL 60612.