Background: - : Chronic rhinosinusitis (CRS) is a common inflammatory disease of the nose and paranasal sinuses that substantially impairs quality of life. Functional endoscopic sinus surgery (FESS) is the standard surgical treatment for patients who fail appropriate medical therapy. Aims and objectives: To evaluate the symptomatic, endoscopic and quality-of-life outcome of FESS in CRS, to correlate disease stage with outcome, to identify factors influencing outcome, and to document complications. Materials and methods: A prospective, single-centre, observational study was carried out on 100 adult patients with CRS who had failed at least 12 weeks of maximal medical therapy and underwent FESS between July 2014 and June 2016 at a tertiary care teaching hospital. Disease was staged by the Lund–Mackay CT score and the Lund–Kennedy endoscopic score. Symptoms were measured with the 22-item Sino-Nasal Outcome Test (SNOT-22) and a 10-point visual analogue scale (VAS) before surgery and at 1, 3 and 6 months afterwards. Data were analysed by repeated-measures ANOVA, paired t-test, chi-square and Fisher’s exact tests; p < 0.05 was considered significant. Results: The mean age was 34.6 ± 11.2 years with a male-to-female ratio of 1.4:1; 34% of patients had nasal polyps. Nasal obstruction (94%) and nasal discharge (88%) were the commonest symptoms. The mean SNOT-22 score fell from 52.4 ± 14.6 pre-operatively to 15.2 ± 8.6 at 6 months (a 71.0% reduction, p < 0.001) and the mean Lund–Kennedy score from 9.6 ± 3.1 to 2.6 ± 1.8 (p < 0.001). The minimal clinically important difference in SNOT-22 was achieved by 91% of patients and the overall success rate (cured or significantly improved) was 79%. Success was lower in patients with asthma (57.1% vs 82.6%, p = 0.030), smokers (58.8% vs 83.1%, p = 0.025) and those with a Lund–Mackay score above 16 (63.3% vs 85.7%, p = 0.012). Minor complications occurred in 28% of patients; there were no major complications. Conclusion: FESS produces significant and clinically meaningful improvement in symptoms, quality of life and endoscopic disease in patients with CRS refractory to medical therapy, with a low complication rate. Asthma, smoking and extensive CT disease predict a poorer result and these patients need careful counselling and intensive postoperative care.
Chronic rhinosinusitis (CRS) is defined as inflammation of the nose and paranasal sinuses lasting twelve weeks or longer, characterised by two or more symptoms, one of which must be nasal blockage or obstruction or nasal discharge, with or without facial pain or pressure and a reduction or loss of smell, together with objective evidence of mucosal inflammation on nasal endoscopy or computed tomography (CT) [1-4]. It is one of the commonest chronic medical conditions in the world, with a prevalence of roughly 5–15% of the adult population in Western countries [1,5]. Reliable community-based data from India are limited, but chronic sinus disease accounts for a considerable share of the outpatient workload in Indian otorhinolaryngology departments [6,7].
The disease imposes a heavy burden on the individual and on society. Patients experience nasal obstruction, discharge, facial pain, hyposmia, disturbed sleep, fatigue and emotional distress, and the impact on health-related quality of life is comparable to that of other major chronic illnesses [8,9]. The direct and indirect economic costs, arising from physician visits, drugs, surgery and loss of productivity, are considerable [10,11].
The aetiopathogenesis of CRS is multifactorial. Obstruction of the osteomeatal complex (OMC) and the other sinus drainage pathways, impaired mucociliary clearance, allergy, bacterial colonisation with superantigen-producing Staphylococcus aureus, biofilms, anatomical variations such as concha bullosa, agger nasi cells and septal deviation, and environmental irritants all contribute [12-14]. In the Indian setting, the hot and humid climate, dust and air pollution, a high burden of allergic disease and the relatively frequent occurrence of fungal rhinosinusitis add further complexity to the clinical picture [15,16].
Current guidelines recommend that the initial treatment of CRS should be medical, consisting of intranasal corticosteroids, saline irrigation and, where indicated, short courses of antibiotics or oral steroids. Surgery is reserved for patients who continue to have significant symptoms despite adequate medical treatment [17-19]. The modern surgical approach is based on the work of Messerklinger in Austria, who demonstrated by endoscopy that the pathology of most sinus disease begins in the narrow clefts of the lateral nasal wall, and who showed that mucociliary clearance of the sinuses is directed towards natural ostia [20,21]. Stammberger and Kennedy introduced the principles of functional endoscopic sinus surgery (FESS) to the wider surgical community, and Wigand developed parallel techniques [22-25]. The philosophy of FESS is to remove the diseased tissue and obstructing bony partitions in the key drainage areas, restoring ventilation and mucociliary clearance while preserving normal mucosa and minimising morbidity.
Numerous studies, including systematic reviews, have documented symptomatic improvement in the great majority of patients after FESS [26-29], and a prospective multi-institutional study has shown superior quality-of-life gains in patients who choose surgery after failed medical management [30]. However, reported success rates vary widely according to the population studied, the definition of “success”, the tools used and the length of follow-up. Prospective data from Indian patients, using validated patient-reported outcome measures together with objective staging, remain relatively scarce, and the spectrum of disease in India (including nasal polyposis with a distinct tissue-eosinophilia profile and a higher fungal burden) may influence outcome [31]. This prospective study was therefore undertaken to evaluate the outcome of FESS in CRS patients at our institution and to determine the factors that predict a good or poor result.
Aims and objectives
To assess the outcome of FESS in patients with CRS in terms of symptom relief, quality of life and endoscopic appearance at 1, 3 and 6 months. To correlate pre-operative CT and endoscopic staging with symptom severity and surgical outcome. To identify patient and disease-related factors that influence outcome. To record the intra-operative and postoperative complications of FESS.
Study design and setting
This was a prospective, single-centre, observational study conducted in the Department of Otorhinolaryngology and Head & Neck Surgery of a tertiary care teaching hospital over a period of two years, from July 2014 to June 2016, with the last patient completing follow-up at the end of the study period. Approval was obtained from the Institutional Ethics Committee and written informed consent was taken from every participant after explaining the nature of the study, the surgical procedure and its risks.
Sample size
Assuming an expected success rate of FESS of 85% on the basis of earlier reports [26,28,29], an absolute precision of 7% and a 95% confidence level, the sample size was calculated from the formula n = Z²pq/d², which gave approximately 100 patients. Consecutive eligible patients were enrolled until this number was reached.
Inclusion criteria
Exclusion criteria
Pre-operative evaluation
A detailed history was recorded on a structured proforma that included duration and nature of symptoms, previous treatment, history of allergy and asthma, smoking, and associated diseases. Anterior rhinoscopy and diagnostic nasal endoscopy were performed with a 4-mm 0° and 30° rigid Hopkins endoscope after topical decongestion and anaesthesia. The endoscopic findings were scored according to the Lund–Kennedy system, which grades polyps, discharge, oedema, scarring/adhesions and crusting on each side, giving a total of 0–20 [32,33]. Non-contrast CT of the nose and paranasal sinuses in coronal and axial planes (3-mm sections) was done in all patients after the medical treatment, and was staged with the Lund–Mackay system, which scores each sinus group from 0 (no opacity) to 2 (total opacity) and the OMC as 0 or 2, to a maximum of 24 [34-36]. Anatomical variations were also recorded using standard terminology [14]. Routine blood counts, blood sugar and a pre-anaesthetic check-up were completed in all patients.
Patient-reported symptom burden was assessed using the 22-item Sino-Nasal Outcome Test (SNOT-22; total score 0–110, higher scores indicating worse disease), which is a validated instrument developed from the earlier SNOT-20 [9,37]. Patients also rated overall symptom severity on a 10-point VAS (0 = no problem, 10 = worst imaginable). The questionnaire was administered in the language best understood by the patient (English, Hindi or the regional language) with assistance for those unable to read.
Surgical technique
All operations were performed under general anaesthesia with controlled hypotension and nasal packing with 1:10,000 adrenaline-soaked patties by a team of experienced surgeons, using 4-mm 0°, 30° and 70° Hopkins rod endoscopes and the Messerklinger anterior-to-posterior technique [22,24,38]. The extent of surgery was tailored to the disease seen on CT and endoscopy. In every case an uncinectomy and middle meatal antrostomy were done. This was followed, as needed, by anterior ethmoidectomy, posterior ethmoidectomy, frontal recess clearance and sphenoethmoidectomy or sphenoidotomy. Septoplasty and resection of a concha bullosa were performed in the same sitting when these anatomical problems contributed to obstruction. Mucosa was preserved wherever possible and the middle turbinate was conserved. The nose was packed with absorbable haemostatic material or a non-adherent pack that was removed within 24 hours.
Postoperative care and follow-up
All patients received a broad-spectrum oral antibiotic for 7 days, analgesics, saline nasal douching from the first postoperative day and a topical intranasal steroid spray for at least three months. Patients with polyposis received a short tapering course of oral prednisolone. Endoscopic toilet with removal of crusts and clots, and division of early adhesions, was carried out at 1 week, 2 weeks and 1 month, and subsequently as required. Patients were reviewed at 1, 3 and 6 months, when SNOT-22, VAS and the Lund–Kennedy endoscopic score were recorded again, and complications were noted.
Definition of outcome
Outcome at six months was graded as follows: cured (asymptomatic with a normal or near-normal endoscopic appearance), significantly improved (≥ 50% reduction in the SNOT-22 score with minimal residual endoscopic disease), marginally improved (< 50% reduction in the SNOT-22 score) and no improvement or worse. “Success” was defined as the combined cured and significantly improved categories. In addition, the proportion of patients who achieved the minimal clinically important difference (MCID) of 8.9 points in the SNOT-22 was determined [37].
Statistical analysis
Data were entered in Microsoft Excel and analysed with SPSS version 20.0 (IBM Corp., Armonk, NY). Continuous variables are expressed as mean ± standard deviation and categorical variables as number and percentage. Serial scores were compared with repeated-measures ANOVA followed by Bonferroni-corrected post-hoc tests, and pre- versus post-operative scores with the paired t-test. Proportions were compared by the chi-square test or Fisher’s exact test, as applicable, and correlations were assessed by Pearson’s coefficient. A p value of less than 0.05 was taken as statistically significant.
A total of 100 patients who fulfilled the criteria were enrolled and all of them completed the six-month follow-up; no patient was lost to follow-up.
The age of the patients ranged from 18 to 62 years with a mean of 34.6 ± 11.2 years. The largest group (31%) was in the 26–35-year decade, and 78% of the patients were below 45 years. There were 58 males and 42 females (male-to-female ratio 1.4:1) (Table 1).
Table 1: Age and sex distribution of the patients (n = 100)
|
Age group (years) |
Male |
Female |
Total |
% |
|
18–25 |
12 |
10 |
22 |
22.0 |
|
26–35 |
18 |
13 |
31 |
31.0 |
|
36–45 |
14 |
11 |
25 |
25.0 |
|
46–55 |
9 |
6 |
15 |
15.0 |
|
> 55 |
5 |
2 |
7 |
7.0 |
|
Total |
58 |
42 |
100 |
100.0 |
Mean age 34.6 ± 11.2 years; range 18–62 years.
The duration of symptoms was between one and three years in 41% of the patients. Nasal obstruction was the commonest presenting symptom (94%), followed by nasal discharge or postnasal drip (88%) and facial pain or pressure (72%); hyposmia was reported by 54% (Table 2, Figure 1). Deviated nasal septum (52%) and allergic rhinitis (38%) were the commonest associated conditions, and 34% of the patients had nasal polyps (CRS with nasal polyps, CRSwNP), while the remaining 66% had CRS without polyps (CRSsNP) (Table 3).
Table 2: Presenting symptoms (n = 100)
|
Symptom |
Number of patients |
Percentage |
|
Nasal obstruction |
94 |
94.0 |
|
Nasal discharge / postnasal drip |
88 |
88.0 |
|
Facial pain / pressure |
72 |
72.0 |
|
Headache |
66 |
66.0 |
|
Hyposmia / anosmia |
54 |
54.0 |
|
Sneezing |
41 |
41.0 |
|
Cough |
28 |
28.0 |
Patients could have more than one symptom.
Table 3: Duration of symptoms and associated conditions (n = 100)
|
Variable |
Number |
Percentage |
|
Duration of symptoms: < 1 year |
18 |
18.0 |
|
Duration: 1–3 years |
41 |
41.0 |
|
Duration: 3–5 years |
24 |
24.0 |
|
Duration: > 5 years |
17 |
17.0 |
|
Deviated nasal septum |
52 |
52.0 |
|
Allergic rhinitis |
38 |
38.0 |
|
Nasal polyps (CRSwNP) |
34 |
34.0 |
|
Smoking (current / ex-smoker) |
17 |
17.0 |
|
Bronchial asthma |
14 |
14.0 |
|
Diabetes mellitus |
9 |
9.0 |
CRSwNP = chronic rhinosinusitis with nasal polyps. Associated conditions are not mutually exclusive.
Figure 1: Presenting symptoms before surgery and persisting symptoms at 6 months
Mucosal oedema was present in 91% and mucopurulent discharge in the middle meatus in 76% of patients. Polyps were seen in 34 patients, of whom 9 had grade I, 15 grade II and 10 grade III disease. On CT, the maxillary sinus (92%) and anterior ethmoid cells (88%) were most commonly involved, and the OMC was obstructed in 85% (Table 4). Anatomical variations were frequent, the commonest being deviated nasal septum, agger nasi cell and concha bullosa. The mean pre-operative Lund–Mackay CT score was 12.8 ± 4.9 (range 3–23) and was significantly higher in CRSwNP (15.9 ± 4.2) than in CRSsNP (11.2 ± 4.3) (p < 0.001). The mean pre-operative Lund–Kennedy score was 9.6 ± 3.1 (Table 5). Thirty patients (30%) had a Lund–Mackay score above 16. The Lund–Mackay score showed only a weak-to-moderate correlation with the pre-operative SNOT-22 score (r = 0.31, p = 0.002) but correlated better with the Lund–Kennedy score (r = 0.46, p < 0.001).
Table 4: Pre-operative endoscopic and CT findings (n = 100)
|
Finding |
Number |
Percentage |
|
Mucosal oedema / hyperaemia |
91 |
91.0 |
|
Mucopus in middle meatus |
76 |
76.0 |
|
Nasal polyps |
34 |
34.0 |
|
Deviated nasal septum |
52 |
52.0 |
|
Agger nasi cell |
64 |
64.0 |
|
Concha bullosa |
29 |
29.0 |
|
Haller (infraorbital) cell |
11 |
11.0 |
|
Paradoxical middle turbinate |
7 |
7.0 |
|
CT: maxillary sinus opacification |
92 |
92.0 |
|
CT: anterior ethmoid involvement |
88 |
88.0 |
|
CT: OMC obstruction |
85 |
85.0 |
|
CT: posterior ethmoid involvement |
61 |
61.0 |
|
CT: frontal sinus involvement |
47 |
47.0 |
|
CT: sphenoid sinus involvement |
38 |
38.0 |
Table 5: Pre-operative staging scores
|
Score |
All patients (n = 100) |
CRSwNP (n = 34) |
CRSsNP (n = 66) |
p value |
|
Lund–Mackay CT score (0–24) |
12.8 ± 4.9 |
15.9 ± 4.2 |
11.2 ± 4.3 |
< 0.001 |
|
Lund–Kennedy endoscopic score (0–20) |
9.6 ± 3.1 |
11.4 ± 2.7 |
8.7 ± 2.9 |
< 0.001 |
|
Lund–Mackay grade: mild (1–8) / moderate (9–16) / severe (> 16), n |
21 / 49 / 30 |
2 / 14 / 18 |
19 / 35 / 12 |
– |
Values are mean ± SD unless stated. Comparison between CRSwNP and CRSsNP by unpaired t-test.
FESS was performed bilaterally in 71 patients and unilaterally in 29. All patients underwent uncinectomy with middle meatal antrostomy, and the extent of further surgery depended on the disease (Table 6). Septoplasty was combined with FESS in all 52 patients with a deviated septum, and a concha bullosa was addressed by lateral lamellectomy in 29 patients. The mean operating time was 74 ± 21 minutes for unilateral and 118 ± 26 minutes for bilateral procedures, and the mean hospital stay was 2.3 days.
Table 6: Surgical procedures performed (n = 100)
|
Procedure |
Number |
Percentage |
|
Uncinectomy with middle meatal antrostomy |
100 |
100.0 |
|
Anterior ethmoidectomy |
88 |
88.0 |
|
Posterior ethmoidectomy |
61 |
61.0 |
|
Frontal recess clearance |
47 |
47.0 |
|
Sphenoidotomy |
38 |
38.0 |
|
Septoplasty |
52 |
52.0 |
|
Concha bullosa lamellectomy |
29 |
29.0 |
Procedures were combined in most patients.
All outcome measures showed a statistically significant improvement over the follow-up period (Table 7, Figure 2). The mean SNOT-22 score fell from 52.4 ± 14.6 before surgery to 28.6 ± 11.8 at one month, 19.7 ± 9.9 at three months and 15.2 ± 8.6 at six months (repeated-measures ANOVA, p < 0.001), a reduction of 71.0% from baseline. The greatest change was seen in the first month, but a further significant fall occurred between 3 and 6 months (p = 0.004). The VAS fell by 68.9% and the Lund–Kennedy endoscopic score by 72.9%. All five SNOT-22 domains improved significantly, the psychological domain showing the largest relative improvement (75.2%) (Table 8).
Table 7: Serial changes in symptom and endoscopic scores (mean ± SD)
|
Parameter |
Pre-op |
1 month |
3 months |
6 months |
p value* |
|
SNOT-22 total score |
52.4 ± 14.6 |
28.6 ± 11.8 |
19.7 ± 9.9 |
15.2 ± 8.6 |
< 0.001 |
|
VAS symptom score (0–10) |
7.4 ± 1.3 |
4.1 ± 1.5 |
2.9 ± 1.4 |
2.3 ± 1.3 |
< 0.001 |
|
Lund–Kennedy score (0–20) |
9.6 ± 3.1 |
5.2 ± 2.4 |
3.4 ± 2.0 |
2.6 ± 1.8 |
< 0.001 |
*Repeated-measures ANOVA. Post-hoc (Bonferroni): each time point differed significantly from pre-operative values (p < 0.001).
Figure 2: Serial SNOT-22 and Lund–Kennedy scores (mean ± SD)
Table 8: SNOT-22 domain scores before and six months after surgery (mean ± SD)
|
SNOT-22 domain |
Pre-op |
6 months |
% reduction |
p value |
|
Rhinologic symptoms |
16.8 ± 5.2 |
4.9 ± 2.9 |
70.8 |
< 0.001 |
|
Extra-nasal rhinologic symptoms |
7.9 ± 3.1 |
2.6 ± 1.8 |
67.1 |
< 0.001 |
|
Ear / facial symptoms |
9.2 ± 3.6 |
2.7 ± 1.9 |
70.7 |
< 0.001 |
|
Psychological dysfunction |
11.3 ± 4.8 |
2.8 ± 2.1 |
75.2 |
< 0.001 |
|
Sleep dysfunction |
7.2 ± 3.0 |
2.2 ± 1.6 |
69.4 |
< 0.001 |
|
Total |
52.4 ± 14.6 |
15.2 ± 8.6 |
71.0 |
< 0.001 |
Paired t-test.
Symptom-wise, nasal obstruction (88.3%), facial pain (87.5%), headache (84.8%) and nasal discharge (84.1%) showed the best resolution, while hyposmia (61.1%) and sneezing (63.4%) improved least (Table 9). Ninety-one patients (91%) achieved the MCID for SNOT-22.
Table 9: Symptom-wise improvement at six months
|
Symptom |
Pre-op (n) |
Persisting at 6 months (n) |
Improvement (%) |
|
Nasal obstruction |
94 |
11 |
88.3 |
|
Nasal discharge / postnasal drip |
88 |
14 |
84.1 |
|
Facial pain / pressure |
72 |
9 |
87.5 |
|
Headache |
66 |
10 |
84.8 |
|
Hyposmia / anosmia |
54 |
21 |
61.1 |
|
Sneezing |
41 |
15 |
63.4 |
|
Cough |
28 |
8 |
71.4 |
At six months 38 patients (38%) were cured, 41 (41%) were significantly improved, 14 (14%) were marginally improved and 7 (7%) showed no improvement, giving an overall success rate of 79% (Table 10, Figure 3). The success rate was 70.6% in CRSwNP and 83.3% in CRSsNP, but this difference did not reach statistical significance (p = 0.14). Polyps recurred in five of the 34 patients with CRSwNP (14.7%) within six months, and four patients (4%) were advised revision surgery.
Table 10: Overall outcome at six months
|
Grade of outcome |
CRSwNP (n = 34) |
CRSsNP (n = 66) |
Total (n = 100) |
% |
|
Cured |
9 |
29 |
38 |
38.0 |
|
Significantly improved |
15 |
26 |
41 |
41.0 |
|
Marginally improved |
6 |
8 |
14 |
14.0 |
|
No improvement / worse |
4 |
3 |
7 |
7.0 |
|
Success (cured + significantly improved) |
24 (70.6%) |
55 (83.3%) |
79 |
79.0 |
Figure 3: Overall outcome at six months (n = 100)
Analysis of factors influencing the result (Table 11) showed that the success rate was significantly lower in patients with bronchial asthma (57.1% vs 82.6%, p = 0.030), in smokers (58.8% vs 83.1%, p = 0.025) and in patients with a Lund–Mackay score above 16 (63.3% vs 85.7%, p = 0.012). A trend towards poorer outcome was noted in diabetics (p = 0.07). Presence of polyps, allergic rhinitis and concomitant septoplasty did not significantly alter the success rate. The six-month SNOT-22 score correlated weakly with the pre-operative Lund–Mackay score (r = 0.38, p < 0.001).
Table 11: Factors influencing success of FESS at six months
|
Factor |
Present (n) |
Success, present n (%) |
Absent (n) |
Success, absent n (%) |
p value |
|
Nasal polyps |
34 |
24 (70.6) |
66 |
55 (83.3) |
0.14 |
|
Bronchial asthma |
14 |
8 (57.1) |
86 |
71 (82.6) |
0.030 |
|
Allergic rhinitis |
38 |
27 (71.1) |
62 |
52 (83.9) |
0.13 |
|
Smoking |
17 |
10 (58.8) |
83 |
69 (83.1) |
0.025 |
|
Diabetes mellitus |
9 |
5 (55.6) |
91 |
74 (81.3) |
0.07† |
|
Septoplasty done |
52 |
44 (84.6) |
48 |
35 (72.9) |
0.15 |
|
Lund–Mackay score > 16 |
30 |
19 (63.3) |
70 |
60 (85.7) |
0.012 |
Chi-square test; †Fisher’s exact test. Differences with p < 0.05 (asthma, smoking, Lund–Mackay score > 16) are significant.
No major complication such as cerebrospinal fluid leak, orbital injury, major haemorrhage requiring transfusion, or visual impairment occurred. Minor complications, comprising 30 events in 28 patients (28%), are listed in Table 12. Crusting persisting beyond four weeks (12%) and synechiae (9%) were the commonest; synechiae were divided in the outpatient clinic under endoscopic guidance in seven patients. Postoperative epistaxis (6%) settled with local measures and re-packing, and periorbital ecchymosis (3%) resolved spontaneously.
Table 12: Complications of FESS (n = 100)
|
Complication |
Number |
Percentage |
|
Prolonged crusting (> 4 weeks) |
12 |
12.0 |
|
Synechiae / adhesions in the middle meatus |
9 |
9.0 |
|
Minor postoperative epistaxis |
6 |
6.0 |
|
Periorbital ecchymosis |
3 |
3.0 |
|
Major complications (CSF leak, orbital, major bleeding) |
0 |
0.0 |
Thirty minor events occurred in 28 patients.
Functional endoscopic sinus surgery has become the standard operation for CRS that has not responded to medical therapy. In this prospective study of 100 patients, FESS produced a significant and sustained reduction in symptoms, an improvement in disease-specific quality of life and a marked improvement in the endoscopic appearance of the sinonasal mucosa over six months, with an overall success rate of 79% and a low rate of minor complications. These findings are in line with the published literature, and add to the limited body of prospective Indian data.
The patients in this study were relatively young, with a mean age of under 35 years and a slight male predominance. A similar demographic profile is commonly seen in Indian ENT practice and may reflect greater occupational and environmental exposure to dust, smoke and pollutants among working-age men, which is an established aggravating factor in chronic sinonasal disease [6,7]. The majority of our patients had suffered symptoms for more than a year before surgery, which emphasises the delay in referral that is frequently observed in our country, where self-medication with decongestants and repeated antibiotic courses is common.
Nasal obstruction and nasal discharge were the commonest complaints, in keeping with the cardinal symptoms that form the basis of the definition of CRS [1,2,8,39]. Hyposmia, present in just over half of our patients, was the symptom that improved the least (61.1%), as other investigators have noted; olfactory dysfunction in CRS is related to mucosal inflammation of the olfactory cleft and to eosinophilic disease, and patients with polyps recover less completely [17,31,40]. Patients should therefore be counselled before surgery that smell may not return fully even when the nose is otherwise clear.
The magnitude of improvement in quality of life that we found is substantial. The mean SNOT-22 score fell by 37.2 points, which is more than four times the MCID of 8.9 points that has been established for this instrument [37], and 91% of our patients crossed that threshold. Comparable improvements in symptom scores and quality of life have been documented in prospective series and in a systematic review of endoscopic sinus surgery [26,27,41]. Senior and colleagues reported that the benefit persists in the long term, with the majority of patients remaining improved after several years [28], and the English national audit showed that improvement in SNOT-22 was sustained at three years [29]. Although our follow-up is shorter, the continuing improvement between the third and sixth month suggests that mucosal healing and remodelling go on for several months, which agrees with the observations of Chambers and colleagues on the time course of recovery after FESS [42].
The endoscopic score fell by almost 73% and showed a similar time course to the symptoms. Although the symptom and endoscopic scores moved in the same direction, the correlations between them are modest. Likewise, the Lund–Mackay CT score in our study correlated only weakly with the SNOT-22 score before surgery. A poor correlation between CT staging and symptoms has been reported repeatedly, because CT records the extent of opacification at one time point and does not capture the severity of symptoms, which are influenced by several other factors such as mucosal sensitivity, mood and comorbidity [36,43]. For this reason, we believe CT should be regarded as a map for surgery and as a prognostic indicator rather than a measure of symptom burden, and that patient-reported outcome measures should be included in every audit of sinus surgery.
The role of surgery relative to medical treatment merits comment. All our patients had failed at least 12 weeks of maximal medical therapy before being offered surgery, which is the policy recommended by current guidelines [1,17]. In a randomised controlled trial, Ragab and co-workers found that medical treatment produced considerable improvement in many patients and that surgery was best reserved for those who fail to respond [18,19]. In a prospective multi-institutional study of patients who elected surgery after failing medical therapy, quality-of-life gains were greater in the surgical group than in those who continued medical treatment [30]. Our results, in a selected group of medical failures, are consistent with these observations.
Patients with nasal polyps had a higher Lund–Mackay and Lund–Kennedy score and a higher SNOT-22 score before surgery, but they improved as much in absolute terms as those without polyps; the proportion of successes was lower (70.6% vs 83.3%), although this was not statistically significant, possibly because of the small subgroup. Earlier studies likewise found that polyp patients present with more severe disease but derive a comparable symptomatic benefit from surgery, with a greater tendency to recurrence [29,40]. We observed recurrent polyps in 14.7% of CRSwNP patients within six months, underscoring the need for long-term topical steroid therapy and regular endoscopic follow-up. In India, eosinophil-predominant polyposis, allergy and fungal colonisation are particularly relevant to the recurrence of polyps, and a thorough histopathological and mycological evaluation of excised tissue is advisable [15,16,31].
Of the factors we analysed, asthma, smoking and extensive disease on CT were significantly associated with a poorer outcome. The influence of asthma has been demonstrated before. Patients with asthma have more extensive sinus disease on CT and higher symptom scores, and although surgery can lessen asthma symptoms and medication use, their sinonasal outcomes are less favourable than in non-asthmatics [44,45]. Smoking is a well-recognised predictor of poor long-term results after FESS, probably because of ciliary dysfunction, persistent mucosal inflammation and impaired wound healing [46]. Other prospective studies have identified the extent of pre-operative disease, allergy and comorbid conditions as determinants of outcome [47-49]. The trend towards a poorer response in diabetic patients in our series (p = 0.07) must be interpreted with caution because of the very small number of diabetics (n = 9), but it is biologically plausible, given the impaired mucosal healing and greater risk of infection in this group. Identifying such patients before surgery allows realistic counselling, optimisation of asthma and glycaemic control, firm advice to stop smoking, and a more intensive postoperative regimen.
The complication rate was low and no major complication occurred. Major complications of FESS, including orbital injury, CSF leak and vascular injury, have been reported in a small percentage of patients in large series, and are linked to anatomical variation, extensive disease with poor visibility, and surgeon inexperience [50,51]. We attribute the absence of major complications to careful study of the CT scan before surgery, with attention to the skull base, the lamina papyracea and the anterior ethmoidal artery, to controlled hypotension that gives a clear operative field, and to the use of the anterior-to-posterior technique with preservation of anatomical landmarks [22,38]. The relatively high frequency of crusting and synechiae is in keeping with the experience of others, and shows that meticulous postoperative endoscopic toilet is as important as the operation itself. Early synechiae were divided in the clinic with no sequelae.
From an Indian perspective, FESS has the advantage of being a day-care or short-stay procedure with fast recovery that does not need external incisions, and its cost is modest compared to the long-term expenditure on repeated antibiotic courses and lost working days [6,10,11]. However, access to endoscopic equipment, CT scanning and structured postoperative care remains uneven across the country, and a large proportion of patients present late. Multicentre registries that use common outcome instruments such as the SNOT-22 would help to define the true benefits of surgery in the Indian population.
Strengths and limitations
The strengths of this study include its prospective design, consecutive enrolment, the use of validated patient-reported and objective scoring systems, a standardised operative technique and postoperative protocol, and complete follow-up. Its limitations are the single-centre design, the absence of a control group treated medically, the short follow-up of six months that does not allow the assessment of long-term recurrence, the subjective nature of some of the outcome measures, and the small size of some of the subgroups (asthma, diabetes) that limits the power of the subgroup comparisons. Olfactory function was recorded as a symptom and not measured with a standardised smell test, and cost-effectiveness was not evaluated. Larger multicentre studies with longer follow-up are needed to confirm these results.
Functional endoscopic sinus surgery is an effective and safe treatment for patients with chronic rhinosinusitis who have not responded to adequate medical therapy. In this prospective study, it produced a 71% reduction in the SNOT-22 score, a 73% reduction in the endoscopic score and an overall success rate of 79% at six months, with only minor complications. Hyposmia responds less well than other symptoms. Asthma, smoking and extensive disease on CT were associated with poorer outcomes, and patients with these features require careful counselling, optimisation of comorbid disease and a vigilant postoperative follow-up. Careful patient selection, meticulous surgical technique and diligent postoperative endoscopic care are the keys to a good result.