Endoscopic sinus surgery for the excision of nasal

Endoscopic sinus surgery for the excision of nasal polyps: A systematic review of safety and effectiveness Kim Dalziel, B.Health.Sc.(Hons.), M.B.,* Ken Stein,* Ali Round, M.B.,* Ruth Garside, M.A.,* and Pam Royle, Ph.D.# (United Kingdom) ABSTRACT Background: Functional endoscopic sinus surgery (FESS) has been used for ⬎20 years for the management of sinus disease including the excision of nasal polyps. Our objective was to perform a systematic review of safety and effectiveness of FESS for the removal of nasal polyps. Methods: The Cochrane Library, MEDLINE, Embase, Science Citation Index, other databases, and websites were searched in January and December 2005 using key words for nasal polyps and endoscopic surgery. All randomized controlled trials, nonrandomized comparative studies, and case series studies that described outcomes associated with FESS for the excision of nasal polyps were included. Forty-two publications were included from the 632 (6.6%) articles initially identified. Two reviewers assessed validity of included studies and extracted relevant data. Results: Three randomized controlled trials, 4 nonrandomized comparative studies, and 35 case series studies were included in the review. FESS was compared with endoscopic polypectomy, Caldwell-Luc, radical nasalization, and intranasal ethmoidectomy. In general, studies were of poor quality and lacked description of important variables influencing surgical outcome. Overall complications for FESS from case series studies ranged from 0.3 to 22.4% (median, 7.0%). Major complications ranged from 0 to 1.5% (median, 0%) and minor complications ranged from 1.1 to 20.8% (median, 7.5%). The potentially most serious complications were cerebrospinal fluid leaks, injury to the internal carotid artery, dural exposure, meningitis, bleeding requiring transfusion, periorbital/orbital fat exposure, and orbital penetration. Symptomatic improvement ranged from 78 to 88% for FESS compared with 43 to 84% for comparative procedures. From case series, symptomatic improvement ranged from 40 to 98% (median, 88%). Conclusion: FESS may offer some advantages in safety and effectiveness over comparative techniques, but wide variation in reported results and methodological shortcomings of studies limit the certainty of these conclusions. Wide variation in complication rates suggests the need for audit of existing practice. Additional high-quality studies with a fuller description of potential confounding factors and effect modifiers will help to define the effectiveness of FESS more clearly. (Am J Rhinol 20, 506 –519, 2006; doi: 10.2500/ajr.2006.20.2923)

F

unctional endoscopic sinus surgery (FESS) has been used for ⬎20 years in the treatment of sinus disease. Definitions of FESS vary, but for the purpose of this review it is defined as a minimally invasive technique, using an endoscope to restore nasociliary clearance of mucous, drainage, and aeration of the sinuses. Advantages are claimed over surgery without an endoscope including a better view of the surgical field, a more precise and thorough clearance of inflamed tissue, fewer complications, and lower recurrence rates.1 The lifetime prevalence of nasal polyps has been estimated as 0.2–1% in the United Kingdom.2 Although surgery is a second-line option, many cases fail to respond to medical management. Reports of complications from FESS have increased as the number of procedures performed has grown.3 However, no comprehensive systematic review of the use of FESS for this common and important condition in surgical rhinology has been published to date. Therefore, the aim of From the *Peninsula Technology Assessment Group, Peninsula Medical School, Exeter, United Kingdom, and #Southampton Health Technology Assessment Centre, University of Southampton, United Kingdom Funded by the National Health Service (NHS) Research and Development (R & D) Health Technology Assessment Programme (Grant 01/34/01) The authors had no conflicts of interest Address correspondence and reprint requests to Dr. Ken Stein, Peninsula Technology Assessment Group, Peninsula Medical School, Southernhay East, Exeter EX1 1PQ, United Kingdom E-mail address: [email protected]

506

this systematic review was to identify and assess available primary research on the safety and effectiveness of FESS for the removal of nasal polyps.

MATERIALS AND METHODS Searches were performed in the following databases: The Cochrane Library (issue 4, 2004), MEDLINE (1966–2005), Embase (1980–2005), and Science Citation Index (1981–2004). In addition, the following databases and websites were searched: Web of Science Proceedings, BIOSIS, CINAHL, National Research Register, Health Management Information Consortium, British Library Catalog, Clinical trials.gov, SERNIP, and UK Medical Devices Agency. Hand searching of reference lists was performed also. Searching was conducted in January 2005 using a range of terms for nasal polyps and endoscopic surgery. Articles not published in English were excluded from the search. Searches of Medline and Embase were updated in December 2005. Complete details of search strategies are available from the authors. All randomized controlled trials and nonrandomized comparative studies that compared any type of FESS with any other surgical intervention for the excision of nasal polyps were included. In addition, we included case series studies including ⬎50 patients with nasal polyps. This limit on number was chosen to maximize representativeness and therefore generalizability. Studies were excluded if they did not report complications or patient-relevant outcomes, focused on the use of the endoscope for diagnosis rather than treatment or

September–October 2006, Vol. 20, No. 5

were a duplicate publication. Study selection was performed independently by two reviewers and differences were resolved by consensus. Using a structured form, the patient characteristics, complications, clinical outcomes, and validity were extracted and assessed by two reviewers. Differences were resolved by consensus. Meta-analysis was considered but not performed because of clinical and methodological heterogeneity between studies.

RESULTS Study Characteristics The search identified 632 articles, 42 (6.6%) of which were included after completing the selection process (Fig. 1). Three studies were randomized controlled trials, 4 studies were nonrandomized comparative studies, and 35 studies were case series. Update searches of Medline and Embase performed in December 2005 yielded no additional studies. The study characteristics are summarized in Tables 1 and 2.

variation in intervention applied (seven studies), loss to follow-up (five studies), and lack of intention to treat analysis (four studies). Case Series Studies. The 32 case series studies lack control groups and are prone to selection bias. Ten studies had prospective study design4–14 and only 10 enrolled consecutive patients, which, theoretically, may reduce selection bias.5,7,8,10,15–20 There was variation in the intervention and concurrent treatments applied within studies. It was unclear from study reports if all patients entered studies at similar points in disease progression, and this is a very important factor in disease persistence and recurrence. Fifteen studies reported no loss to follow-up.5,11–13,15,17,19–27 None of the remaining studies analyzed results on an intention to treat basis. There were no specific reports of the methods used to assess complications although the majority of studies (32/42, 76%) reported complications. Fifteen of the 32 case series assessed the primary outcome (symptomatic improvement) objectively or with blinding.5,7–9,11,13,16,22,25,27–32 None of the studies reported using a validated disease-specific instrument for symptom evaluation.

Quality Assessment

Safety of FESS

Comparative Studies. Overall, study quality was poor (Table 3). The main threats to internal validity were inadequate approach to random sequence generation and concealment (seven studies), lack of blinding (seven studies), baseline differences between study groups (six studies),

Comparative Studies. Three33–35 of the seven comparative studies reported complication rates. Harkness and colleagues34 reported a complication rate of 1.4% for FESS compared with 0.8% for conventional procedures. Jankowski and colleagues33 reported no complications for the functional eth-

Figure 1. Inclusion process.

American Journal of Rhinology

507

Table 1 Characteristics of included comparative studies Author/Date

Kurent and Zargi, 199847 Penttilä et al., 199735

Venkatachalam and Bhat, 199848 Harkness et al., 199734

Jankowski et al., 199733

Ünlü et al., 199449

Fawehinmi et al., 200351

No. of Patients (No. with Polyps) 40 (40)

Country

Slovenia

Intervention

Comparative Procedure

Endoscopic ethmoidectomy FESS

Endoscopic polypectomy

150 (97)

Finland

50 (50)

New Delhi

FESS

Conventional procedures

U.K.

FESS

Conventional surgery (sphenoid drainage, BAWO, frontal drainage, polypectomy, external approaches, BINA)

76 (76)

France

Functional ethmoidectomy

Radical nasalization

100 (22)

Turkey

ESS

Caldwell-Luc

Saudi Arabia

FESS

Caldwell-Luc

2523 (unclear)

32 (32)

Caldwell-Luc

Median (range) 76 (32–2523) BAWO ⫽ bilateral antral washout; BINA ⫽ bilateral intranasal antrostomy; Max ⫽ maximum.

moidectomy group and 7.7% for the radical nasalization group. Penttila¨ and colleagues35 reported that there were no major complications in either the FESS or the Caldwell-Luc groups. Case Series Studies. Reported complications ranged from 0.3 to 22.4% (median, 7.0%), with much variation in severity reported. Major complications ranged from 0 to 1.5% (median, 0%),6,11,15,16,21,29,30,36–40 and minor complications ranged from 1.1 to 20.8% (median, 7.5%).15–17,21,36–40

Types of Complication Hemorrhage. Figure 2 presents the median percentage across studies (comparative and case series) for each type of hemorrhagic complication associated with FESS. Median

508

reports of “general bleeding” or severe bleeding/bleeding requiring packing or admission were 2.5 and 2.2%, respectively. Infection. Postoperative infection was reported for 16% of FESS procedures and 28% of conventional procedures in the randomized trial by Venkatachalam and Bhat.40 Sobol and colleagues41 reported 0.3% meningitis and 0.5% orbital cellulitis for FESS. Wigand and Hoseman37 reported 0.5% meningitis for FESS. Garrel and colleagues32 reported 1.6% methicillin-resistant Staphylococcus aureus infection. Intranasal Complications. Figure 3 illustrates the median percentage across studies (comparative and case series) for each type of intranasal complication for FESS. The most common intranasal complications were synechia formation


Table 1 Continued Surgical Indications

Postoperative Medical Management

Massive bilateral Topical steroids polyposis Rhinogenous Not stated chronic maxillary sinusitis

Nasal polyposis

Steroid nasal spray (budesonide) 2 puffs twice a day for 6 mon Not stated

Chronic rhinosinusitis (46%), polyposis (37%), recurrent sinusitis (10%), miscellaneous (7%) Diffuse nasal Steroid spray polyposis (600 mg beclomethasone/ day) and a single dose of depot corticosteroid intramuscularly Chronic/recurring Not stated acute rhinosinusitis Fungal sinusitis associated with nasal polyps

Main Outcome Measures

Polyp recurrence

Average Age % Male in Years (range) —

—

Symptomatic improvement, revision surgery, patency, and complications Symptomatic improvement and complications

48 (14–88)

44

25

Max 108

(20–39)

70

—

17 (6–30)

Symptomatic improvement and complications

45 (2–89)

56

—

Max 6

Symptomatic improvement, revision surgery, and complications

46 (26–71)

70

62

29 (18–36)

38 (18–68)

52

–

—

39 (16–62)

75

—

12

Symptomatic improvement, recurrence, and patency Steroid sprays for 12 Disease recurrence mon

45 (38–48)

(median, 10.4%) and middle meatal stenosis/closure (median, 6.7%). Orbital/Ocular Complications. Figure 4 presents the median across studies for reported orbital/ocular complications associated with FESS. The most frequent orbital complications were periorbital/orbital fat exposure (median, 2.1%), orbital penetration (2.1%), and lesion of periorbit (1.4%), although the latter was only reported by one study. Pharyngeal/Mouth Complications. Numbness of teeth, lips, or cheeks was reported for 3% of the FESS group compared with 49% of the Caldwell-Luc group by Penttila¨ and colleagues.35 Harkness and colleagues34 reported 0.2% of cases with numbness of teeth or lips in the FESS group compared with no cases in the conventional procedures group. The same study reported 0.09% of cases with sphenopalatine adhesions and 0.09% of cases with palatal ulceration in the FESS group (but


Previous Mean Length Surgery of Follow-Up (%) in Months (range) — —

63 (44–75) 44 (25–62)

Max 33

not for conventional procedures). Weber and colleagues18 reported that 0.3% of people who underwent FESS experienced numbness of teeth or upper lips. Intracranial Complications. Fourteen case series studies4,15,16,18,22–25,30–32,36,37,41 reported cerebrospinal fluid (CSF) leakage that ranged from 0 to 2.3% (median, 0.3%). Dural exposure was 0.2% of cases in the Jiang study31 and injury to the internal carotid artery was 0.3% of cases in the Weber study.18 Systemic Complications. No systemic complications were reported by the comparative studies. Mild bronchospasm was reported in the Katsantonis,39 Kennedy,29 and Massegur25 studies with rates of 0.4, 0.8, and 0.8%, respectively. Severe bronchospasm was reported at a rate of 1.5% by Garrel and colleagues.32 Cardiac arrest occurred in 0.4% of cases in one study25 and cardiac ischemia was 1.3% of cases in another study.22

509

510

September–October 2006, Vol. 20, No. 5 Taiwan

1112 (499)

Jiang et al., 200131

U.S.

Denmark

112 (51)

U.S.

200 (68)

Jakobsen and 237 (146) Svendstrup, 200010

Jacobs et al., 199738

Norway

Netherlands

230 (92)

72 (72)

Stoop et al, 199212

Germany and India

Germany

325 (325)

Weber et al., 199718

U.S., France, Canada

Country

220 (220)

50 (50)

Klossek et al., 199711

Wigand and Hosemann, 198937 Danielsen and Olofsson, 19969 Friedman et al., 200015

No. of Patients (No. with Polyps)

Author/ Date

ESS opening at least one sinus

FESS Stammberger approach

Endoscopic frontal sinus surgery Endoscopic ethmoidectomy modified Messerklinger approach

ESS Messerklinger technique

Endoscopic ethmoidectomy

ESS

Endonasal microendoscopic pansinusoperation

Total sphenoethmoidectomy

Intervention

Table 2 Characteristics of included case series studies

Persistent chronic sinusitis 100% (Polyposis 34%) Diffuse sinonasal polyposis 27%, middle meatal polyposis 19%, encroaching cells 19%, hyperplastic sinusitis 13%, frontal recess stenosis 9%, frontal mucocele 4% Chronic infections sinusitis 38%, nasal polyps/ polyposis 62% Chronic sinusitis 51%, nasal polyps 46%, antrochoanal polyps 3%

Sinusitis 39%, nasal polyposis 40%

Nasal polyposis involving all sinuses 100% Bilateral chronic polypoid ethmoid sinusitis 100% Nasal polyps (extensive polyposis 44%) Nasal polyposis 100%

Surgical Indications

Not stated

Not stated

Steroid nasal inhaler for 5 days

Not stated

60% Received steroid treatment

Not stated

2 ⫻ 50 ␮g of budesonide for several months

Postoperative Interventions Used

(5–84)

46 (11–92)

(17–72)

41 (14–76)

(2–79)

—

44 (16–72)

—

47 (18–66)

Average Age in Years (range)

62

59

—

—

19

—

—

—

54

% Male

—

10

29

20

20

—

—

—

—

Previous Surgery (%)

(7–126)

12

16 (6–42)

12 (6–31)

41

—

(6–12)

(10–120)

Minimum 36

Mean Length of Follow-Up in Months (range)


511

650 (306)

250 (203)

90 (50)

283 (48– 51% of sides)

Lund and MacKay, 199424

Massegur et al., 199525

Moses et al., 199826 Nishioka et al., 199417

Venkatachalam 210 (66) and Bhat, 199940

393 (185)

U.S.

250 (131)

Levine, 199050

Sobol et al., 199241

U.S.

90 (53)

Lawson et al., 19914

New Delhi

Canada

Spain

U.K.

U.S.

U.S.

U.S.

120 (71)

Kennedy, 199229

U.S.

972 (?)

Katsantonis et al., 199439

FESS modified Messerklinger approach and partial middle turbinectomy FESS using standard technique, including anterior and posterior ethmoidectomies FESS Messerklinger approach

Not stated

Beclomethasone or budesonide spray 2 times daily

Intranasal steroids, oral steroid, and antibiotics

Aqueous beclomethasone nasal spray

Not stated

Existing oral prednisone had treatment tapered

Not stated

Chronic hyperplastic sinusitis 100% (ethmoidal polyps 13%, antrochoanal polyps 19%)

Not stated

Chronic sinusitis 100% Not stated (polyposis 47%)

Chronic sinusitis 100% Intranasal steroids for at least 6 mon

Sphenoethmoidectomy Chronic hyperplastic modified Yankauer rhinosinusitis, approach recurrent nasal (marsupialization) polyposis, sinobronchial syndrome, recurrent purulent pansinusitis ESS Nonpolypoid disease 41% middle meatal polyposis 31%, diffuse polyposis 28% ESS modified Focal 23%, pansinusitis Yankauer 10%, panpolyposis technique 27%, asthma/polyps 32%, asthma/ sinusitis 8% FESS modified Sinonasal polyposis Messerklinger 52%, chronic approach some sinusitis 48% with laser ESS (functional for Chronic rhinosinusitis chronic and acute 51%, gross polyposis sinusitis) 47%, acute recurrent rhinosinusitis 2% ESS modified Antrochoanal polyps Messerklinger and 12%, polyposis/ASA Stammberger sensitivity 25%, technique polyposis/non ASA 45%, chronic suppurative sinusitis 14%, mucoceles 3%, schwannoma 0.4%, aspergillomas 2% ESS (revision) Polyposis 56%

31–40 (7–66)

45 (17–77)

44 (4–83)

42 (20–77)

—

—

—

49 (21–81)

(15–77)

—

58

50

57

40

—

—

—

67

—

—

19

32

—

100

23

—

5

40

71

—

18 (9–33)

Maximum 12

15 (1–44)

23 (8–44)

(12–48)

6

17 (12–42)

42 (2–14)

18 (3–51)

14 (6–30)

512


500 (136) U.S.

Friedman et al., 200021

509 (139) U.S.

500 (246) Austria

Stammberger and Posawatz, 199028

U.S.

Germany

Schaefer, 199823

79 (58)

115 (89)

Delank and Stoll, 199813

Park et al., 199822

200 (147) U.S.

Davis et al., 199116

Austria

124 (53)

Wolf et al., 199530

Country

No. of Patients (No. with Polyps)

Author/Date

Table 2 Continued Surgical Indications

Chronic recurring sinusitis 57%, chronic recurring sinusitis/ diffuse polyps/ antrochoanal polyps 42% FESS Messerklinger Chronic sinusitis approach 26%, chronic sinusitis and polyps 74% FESS Wigand approach No polyps 23%, Polyps near middle turbinate 25%, Polyps within upper/middle turbinate 40%, Total blockage by polyps 12% ESS and middle turbinate Persistent chronic medialization sinusitis 73%, (microdebrider) polyposis 27% FESS removing uncinate Chronic sinusitis process, bulla ethmoidalis, or recurrent and anterior ethmoid cells acute sinusitis 100% (Nasal polyposis 73%) Endoscopic intranasal Symptomatic ethmoidectomy and middle sinusitis 63%, meatus antrostomy/ polyps sphenoidotomy/frontal ⫾ sinusitis 27%, sinusotomy orbital abscess/ sinusitis 1%, thyroid eye disease 8%, AFS 2% FESS Messerklinger Massive nasal approach polyposis 49%

ESS Messerklinger and Stammberger approach

Intervention

12 (3–16)

Average Age in Years (range)

Topical cortisone in special cases

Not stated

Systemic steroids for 2 days, antibiotics

Not stated

—

42 (12–84)

50 (6–81)

—

Topical steroids, (4–83) antibiotics, steroid nasal spray for ⬍3 mon Not stated 44 (14–79)

Antibiotics and topical corticosteroids

Postoperative Interventions Used

—

51

—

—

55

—

48

—

50

56

0

0

—

—

% Previous Male Surgery (%)

Maximum 120

(1–60)

19 (12–108)

10 (6–18)

—

Maximum 36

—

Mean Length of Follow-Up in Months (range)


513

Romania

227 (227)

Albu et al., 20036

Endonasal sinus Chronic sinusitis and Not stated surgery with severe polyposis 76%, endoscopic Chronic ethmoiditis control (including mostly severe maxillary sinus or polyposis 24% ethmoidectomy) Wigand technique ESS Severe sinonasal Inhaled steroids for polyposis 100% 3–6 wk, decongestants, antihistamines, and immunotherapy as needed ESS Messerklinger Chronic rhinosinusitis Antibiotics 2 wk, technique 100% (nasal polyps inhaled nasal 28%) steroids and Guaifenesin 1 yr FESS modified Nasal polyposis Antihistamine, local Messerklinger (antrochoanal polyps steroids for 3 mon 53%, ethmoidal polyps 47%), ESS Messerklinger chronic rhinosinusitis Oral prednisolone, technique 100% antibiotics 2 wk, and/or nasal steroids 12 wk Endonasal Sinonasal polyposis Antibiotics and oral ethmoidectomy 100% corticosteroids 7 days, topical nasal steroids ESS using Bilateral nasal polyps Topical corticosteroids Messerklinger, 100% in required Stammberger and Kennedy techniques and May modifications SphenoethNasal polyposis 100% Topical steroids, oral moidectomy (isolated polyposis steroids, and 48%) antibiotic if needed FESS Chronic rhinosinusitis Not stated 100% (polyps present 39%)

AFS ⫽ allergic fungal sinusitis; ASA ⫽ aspirin sensitive asthma

201 (50–1112)

U.S.

France

132 (132)

Garrel et al., 200332

201 (78)

Turkey

130 (72)

Dursun et al., 200336

Deal and Kountakis, 20048 Median (range)

Bangladesh

100 (100)

Haque et al., 200420

France

U.S.

178 (50)

Perry and Kountakis, 20035

64 (64)

U.S.

118 (118)

Wynn and HarEl, 200427

Dufour et al., 20047

Germany

315 (?)

Wigand et al., 197819

52

70

67

64

49

56

51

45

—

—

38

36

50

27

0

—

43

—

45 (12–50) 55 (19–70) 29 (0–100)

49 (18–80)

45 (9–77)

46 (19–67)

46

36 (9–70)

30 (7–45)

47

44

—

18 (6–78)

Minimum 12

78 (60–96)

24 (12–60)

60

60 (41–63)

Maximum 12

Minimum 12

40 (12–168)

(3–6)

514


Penttila¨ et al., 199735

RCT

⫻ ⫻ ⫻ ? ? 0 ⫻ NA ⫻ ⫻ ⫻

NA ⫻

⫻ ⫻ ⻫ ? ? 23% ⫻ ⫻ ⫻ ⫻ ⫻

NA ⫻

Nonrandomized Nonrandomized RCT Nonrandomized Nonrandomized Comparative Comparative Comparative Comparative Study Study Study Study Venkatachalam Harkness et al., Jankowski et al., Kurent and Ünlü et al., Fawehinmi et and Bhat, 199734 199733 Zargi, 199449 al., 200351 40 47 1998 1998 50 (50) 2523 (946) 76 (76) 40 (40) 100 (22) 32 (32)

RCT

No. of patients (no. with 150 (97) polyps) Adequate randomization to ⫻ ? ⫻ ⫻ ⫻ groups? Concealment of allocation? ? ? ⫻ ⫻ ? Exclusion criteria specified? ⻫ ⫻ ⫻ ⫻ ⫻ ? ? ? ⫻ ⻫ Comparability of groups at baseline? Same intervention to all ? ⫻ ? ? ? patients? Percentage loss to follow-up 15% 4% 0 18% 13% Sample size calculation? ⫻ ⫻ ⫻ ⫻ ⫻ Analysis performed on ITT ⫻ ⻫ NA ⫻ ⫻ basis? Blinding of patients/ ⫻ ⫻ ⫻ ⫻ ⫻ providers of care? Blinded/independent main ⫻ ? ⫻ ⻫ ⫻ outcome measurement? Point estimates and measure ⫻ ⫻ ⫻ ⻫ ⫻ of variation reported for main outcome? Intercenter variability NA NA ⫻ NA NA assessed? Conflicts of interest stated? ⫻ ⫻ ⻫ ⫻ ⫻ ⻫ ⫽ yes, ⫻ ⫽ no, ? ⫽ uncertain, ITT ⫽ intention to treat; NA ⫽ not applicable, RCT ⫽ randomized controlled trial.

Author/Date

Study Design

Table 3 Methodological characteristics of included comparative studies

Figure 2. Median percentage of FESS complications for the types of hemorrhage reported in the included studies.

Figure 3. Median percentage of FESS complications for the types of intranasal complications reported in the included studies. Nonspecific or Other Complications. Cheek pain/tenderness occurred in 4% of cases for the FESS group and 76% of cases for the Caldwell-Luc group in the Penttila¨ study.35 Pain was reported for 0.7% of FESS procedures in the Jiang study.31 Unremoved nasal packs or sponges occurred in 0.08 and 0.2% of cases (Jiang31 and Stammberger28 studies, respectively). Repeat surgery to complete the FESS procedure was required for 0.2% of people in the study by Stammberger and Posawatz.28 Harkness and colleagues34 reported that 0.07% of the conventional procedures group had cheek edema, 0.07% had unspecified complications, and 0.07% had anosmia. Jiang and colleagues31 reported atrophic rhinitis for 0.08% of procedures and Stammberger and Posawatz reported soft tissue infiltration for 1.8% of cases.28


Relationship between Complications and Revision Surgery. We explored the relationship between complication rates and revision surgery because this may carry greater risks due to prior removal of sinus landmarks (Fig. 5). There was no significant relationship.

Clinical Effectiveness Factors Influencing Surgical Success. Information regarding medical management, stage of disease, instrument used to measure outcomes, length of follow-up, and physician experience are all critical to determining success of sinus surgery. The current studies provide limited information (see Table 2) on these factors and results should be interpreted with caution.

515

Figure 4. Median percentage of FESS complications for the types of orbital/ocular complications reported in the included studies.

DISCUSSION

Figure 5. Relationship between rate of complications and revision surgery.

Symptomatic Improvement. Symptomatic improvement after FESS ranged from 78 to 88% versus 43 to 84% for comparative procedures (statistically significant difference in two studies, Fig. 6). In the case series, symptomatic improvement ranged from 40 to 98% with a median of 88%. The relationship between symptomatic improvement and length of follow-up is shown in Fig. 7. Disease Recurrence/Revision. Disease recurrence ranged from 4 to 60% with a median of 20% across all studies. Similarly, for revision surgery the range was 3–42% with a median of 6%. Sense of Smell. Preoperative smell disturbance ranged from 16 to 100% with a median of 47% compared with postoperative ranges from 3 to 38% with a median of 16% (follow-up ranged from an average of 3–60 months). Nasal Obstruction and Patency. Nasal obstruction ranged from 32 to 95% preoperatively, with a median of 70% compared with a postoperative range of 0–40% with a median of 8%. Overall postoperative patency of the middle meatus ranged from 57 to 100% with a median of 93% (follow-up ranged from 12 to 60 months).

516

Case series studies of FESS reported a wide range of complication rates from 0.3 to 22.4%. However, the distribution is skewed with a median of 7.0%. Major complications ranged from 0 to 1.5% (median, 0%) and minor complications ranged from 1.1 to 20.8% (median, 7.5%). Most frequent complications were hemorrhage (11 studies), CSF leaks/rhinorrhea (14 studies), and periorbital/orbital fat exposure (8 studies). The most serious reported complications were CSF leaks (median, 0.3%), injury to the internal carotid artery (0.3%), dural exposure (0.2%), meningitis (median, 0.15%), bleeding requiring blood transfusion (median, 0.2%), and orbital penetration (2.1%). The overall complication rates found in this review are a wider range than those reported elsewhere. Stankiewicz estimates an overall complication rate for ESS ranging from 2 to 17%.42,43 Our review reports lower major complication rates (median, 0%) than those published previously. Kennedy and colleagues44 surveyed all U.S. Otolaryngologists and reported 0.4% major complications. Cumberworth and colleagues45 surveyed U.K. Otolaryngologists and found 0.23% major complications. We identified a slightly higher median CSF leak rate (0.15%) than the 0.1% figure reported by Kennedy and colleagues.44 Overall complication rates were higher in FESS than comparative procedures in one study (1.4% versus 0.8%),34 equivalent in one study (0%), and lower in a third (0% versus 7.7%).33 The difference in adverse events is clearly related to the risk associated with the comparative procedure and prone to selection bias between studies. Facial numbness is much less common in FESS as would be expected given the fundamental difference in surgical approach. FESS may be more effective than comparative procedures in terms of symptomatic improvement. Smell disturbance and nasal obstruction were lower postoperatively than before FESS. However, the poor quality of available evidence and the lack of description of other important factors influencing success limit the certainty of these conclusions.


Figure 6. Forest plot showing symptomatic improvement. clearly important in surgical success and disease recurrence and may be more important than the type of surgical intervention. Sinus surgery is a difficult procedure to standardize and in most cases the surgical technique and its success are influenced by the extent of underlying disease (disease staging) and the presence of comorbidities. Operator experience is important in sinus surgery.42,43 However, the case mix of more experienced surgeons may include proportionately more complex cases resulting in poorer outcomes independent of operator experience.45 Other important potential confounders include preoperative therapy and planning, anesthesia, instrumentation, patient positioning, specific operative techniques, and postoperative medical treatment.46 Figure 7. Relationship between symptomatic improvement and length of follow-up.

Limitations of Included Evidence A total of 12,329 patients were enrolled in the included studies of which only 240 (2%) were enrolled in randomized controlled trials and only 2971 (24%) were enrolled in studies with some type of comparison group. Often, it is difficult from a methodological and ethical point of view to design highquality surgical trials, although the fact remains that only limited conclusions can be drawn from case series evidence. Challenges include strong preferences around indication and procedure, the heterogenous nature of the intervention, learning curve effects, and the impossibility of blinding the research team. A number of factors may influence outcomes other than the type of procedure performed, which must be carefully considered. Many of the studies included in this review lacked important information about possible confounding variables and effect modifiers. Duration of follow-up has an important effect on the outcome of recurrence and it is unfortunate that many case series have not been maintained longer to allow a clearer picture to emerge of efficacy in this long-term condition. Medical management (pre- and postoperatively) is


Limitations of This Review The scope of this assessment was limited to only one possible application of ESS (nasal polyps). Although we have focused on predominantly polypoidal disease, FESS is used across a spectrum of disease, in conjunction with chronic sinusitis. Studies were only included where at least 50 patients had nasal polyps thus allowing broad generalizations. It was not possible from the reported evidence to systematically separate the results for individual patients who did or did not have nasal polyps It is possible that some case series may have been missed using the search strategy because of difficulty identifying this study design, the high potential for publication bias, and the limitation of English language. However, we do not believe any omissions would have a significant effect on our conclusions. Although we excluded smaller case series, the most realistic estimates of safety effectiveness are likely to come from the larger series. Appraisal of case series studies is difficult because of the lack of information as to which methodological characteristics are most important for validity. The relative importance of characteristics such as prospective design, consecutive selection, and sample size on selection bias is unknown. In highly standardized technologies such as drug treat-

517

ments, the absolute risk of harms may remain reasonably constant as borderline risk for primary clinical outcomes increases. This may not hold for heterogenous interventions such as FESS and precludes statistical pooling across studies because, in more severe cases, the intervention may be more intensive. Despite these limitations, this is the most comprehensive review of the evidence for FESS and highlights the gaps in reported information on this important technology.

4. 5.

6.

7.

Implications for Research and Practice The wide range in complication rates for FESS suggests the need to develop large prospective audits to identify and address potential limitations in practice. There remains a clear need for high-quality, long-term comparative evidence to support use of FESS in nasal polyposis. There is a need to advance debate around suitable methods for establishing effectiveness of surgical interventions given current ethical and methodological challenges. Policy makers and clinicians also will be interested in the overall management of sinus disease (combined medical and surgical) and the cost effectiveness of interventions. Additional research should address these issues. Standardization of procedures is difficult across widely varying cases; although this is ideal, the minimum requirement to enable conclusions to be drawn is a thorough description of what occurred. Outcomes should be objectively and independently measured using validated, disease-specific instruments to limit bias and a particularly useful advance would be agreement over a core set of outcome measures for sinus disease.

CONCLUSION The endoscope has been used for ⬎20 years in sinus surgery and is standard equipment in most institutions. The results of the studies examined suggest that major complications from FESS are relatively rare and there is wide variation in the incidence of mild complications. ESS improves overall and specific symptoms, although there was wide variation in results, making it difficult to provide accurate estimates of the size effects. Variation in outcome and complication rates suggests the need for clearer clinical guidelines and audit in FESS. The quality of existing research is limited and future efforts should concentrate on producing a fuller description of potential confounding factors or effect modifiers to define more clearly the place of FESS in specific populations.

ACKNOWLEDGMENTS The authors gratefully acknowledge the help and support of the following otolaryngologists, who provided clinical and methodological advice throughout the project: Valerie Lund, Ian MacKay, Rob Slack, and Dick Garth.

8.

9.

10.

11.

12.

13.

14. 15.

16.

17.

18.

19.

20.

21.

22.

REFERENCES 1.

Slack R, and Bates G. Functional endoscopic sinus surgery. Am Fam Physician 58:707–718, 1998. 2. Drake-Lee A. Nasal polyps. In Allergic and Non-Allergic Rhinitis. Mygin N, and Naclerio RM (Eds). Copenhagen: Munksgaard, 167–173, 1993. 3. Neuman TR, Turner WJ, and Davidson TM. Complications of endoscopic surgery. Ear Nose Throat J 73:585–590, 1994.

518

23. 24. 25.

Lawson W. The intranasal ethmoidectomy—An experience with 1,077 procedures. Laryngoscope 101:367–371, 1991. Perry BF, and Kountakis SE. Subjective improvement of olfactory function after endoscopic sinus surgery for chronic rhinosinusitis. Am J Otolaryngol 24:366–369, 2003. Albu S, Tomescu E, Mexca Z, et al. Recurrence rates in endonasal surgery for polyposis. Acta Oto-Rhino-Laryngologica Belgica 58:79–86, 2004. Dufour X, Bedier A, Ferrie JC et al. Diffuse nasal polyposis and endonasal endoscopic surgery: Long-term results, a 65case study. Laryngoscope 114:1982–1987, 2004. Deal RT, and Kountakis SE. Significance of nasal polyps in chronic rhinosinusitis: Symptoms and surgical outcomes. Laryngoscope 114:1932–1935, 2004. Danielsen A, and Olofsson J. Endoscopic endonasal sinus surgery. A long-term follow-up study. Acta Otolaryngol 116:611–619, 1996. Jakobsen J, and Svendstrup F. Functional endoscopic sinus surgery in chronic sinusitis—A series of 237 consecutively operated patients. Acta Otolaryngol 120 (suppl 543):158–161, 2000. Klossek JM, Peloquin L, Friedman WH, et al. Diffuse nasal polyposis: Postoperative long-term results after endoscopic sinus surgery and frontal irrigation. Otolaryngol Head Neck Surg 117:355–361, 1997. Stoop AE, van der Heijden HA, Biewenga J, et al. Clinical aspects and distribution of immunologically active cells in the nasal mucosa of patients with nasal polyps after endoscopic sinus surgery and treatment with topical corticosteroids. Eur Arch Otorhinolaryngol 249:313–317, 1992. Delank KW, and Stoll W. Olfactory function after functional endoscopic sinus surgery for chronic sinusitis. Rhinology 36:15–19, 1998. Danielsen A. Functional endoscopic sinus surgery on a day case out-patient basis. Clin Otolaryngol 17:473–477, 1992. Friedman M, Landsberg R, Schults RA, et al. Frontal sinus surgery: Endoscopic technique and preliminary results. Am J Rhinol 14:393–403, 2000. Davis WE, Templer JW, Lamear WR, et al. Middle meatus anstrostomy: Patency rates and risk factors. Otolaryngol Head Neck Surg 104:467–472, 1991. Nishioka GJ, Cook PR, Davis WE, et al. Immunotherapy in patients undergoing functional endoscopic sinus surgery. Otolaryngol Head Neck Surg 110:406–412, 1994. Weber R, Draf W, Keerl R, et al. Endonasal microendoscopic pansinusoperation in chronic sinusitis. II. Results and complications. Am J Otolaryngol 18:247–253, 1997. Wigand ME, Steiner W, and Jaumann MP. Endonasal sinus surgery with endoscopical control: From radical operation to rehabilitation of the mucosa. Endoscopy 10:255–260, 1978. Haque MR, Hossain MM, Kundu SC, et al. A study of functional endoscopic sinus surgery technique. Mymensingh Med J 13:39–42, 2004. Friedman M, Landsberg R, and Tanyeri H. Middle turbinate medialization and preservation in endoscopic sinus surgery. Otolaryngol Head Neck Surg 123:76–80, 2000. Park AH, Lau J, Stankiewicz J, et al. The role of functional endoscopic sinus surgery in asthmatic patients. J Otolaryngol 27:275–280, 1998. Schaefer SD. An anatomic approach to endoscopic intranasal ethmoidectomy. Laryngoscope 108:1628–1634, 1998. Lund VJ, and Mackay IS. Outcome assessment of endoscopic sinus surgery. J R Soc Med 87:70–72, 1994. Massegur H, Adema JM, Lluansi J, et al. Endoscopic sinus surgery in sinusitis. Rhinology 33:89–92, 1995.


26.

27.

28.

29. 30.

31.

32.

33.

34.

35.

36.

37. 38. 39.

Moses RL, Cornetta A, Atkins JPJ, et al. Revision endoscopic sinus surgery: The Thomas Jefferson University experience. Ear Nose Throat J 77:190–202, 1998. Wynn R, and Har-El G. Recurrence rates after endoscopic sinus surgery for massive sinus polyposis. Laryngoscope 114:811–813, 2004. Stammberger H, and Posawetz W. Functional endoscopic sinus surgery. Concept, indications, and results of the Messerklinger technique. Eur Arch Otorhinolaryngol 247: 63–76, 1990. Kennedy DW. Prognostic factors, outcomes and staging in ethmoid sinus surgery. Laryngoscope 102:1–18, 1992. Wolf G, Greistorfer K, and Jebeles JA. The endoscopic endonasal surgical technique in the treatment of chronic recurring sinusitis in children. Rhinology 33:97–103, 1995. Jiang RS, and Hsu CY. Endoscopic sinus surgery for the treatment of chronic sinusitis in geriatric patients. Ear Nose Throat J 80:230–232, 2001. Garrel R, Gardiner Q, Khudjadze M, et al. Endoscopic surgical treatment of nasal polyposis—Medium term outcomes (mean follow up 5 years). Rhinology 41:96, 2003. Jankowski R, Pigret D, and Decroocq F. Comparison of functional results after ethmoidectomy and nasalization for diffuse and severe nasal polyposis. Acta Otolaryngol 117: 601–608, 1997. Harkness P, Brown P, Fowler S, et al. A national audit of sinus surgery. Results of the Royal College of Surgeons of England comparative audit of ENT surgery. Clin Otolaryngol 22:147–151, 1997. Penttila M, Rautiainen M, Pukander J et al. Functional vs. radical maxillary surgery. Failures after functional endoscopic sinus surgery. Acta Otolaryngol Suppl 529:173–176, 1997. Dursun E, Korkmaz H, Eryilmaz A, et al. Clinical predictors of long-term success after endoscopic sinus surgery. Otolaryngol Head Neck Surg 129:526–531, 2003. Wigand ME, and Hosemann W. Microsurgical treatment of recurrent nasal polyposis. Rhinol 27 (suppl 8): 25–29, 1989. Jacobs JB. 100 years of frontal sinus surgery. Laryngoscope 107:1–36, 1997. Katsantonis GP, Friedman WH, and Bruns M. Intranasal sphenoethmoidectomy: An evolution of technique. Otolaryngol Head Neck Surg 111:781–786, 1994.


40.

41.

42. 43. 44.

45.

46.

47.

48.

49.

50.

51.

Venkatachalam VP, and Bhat A. Functional endoscopic sinus surgery—A newer surgical concept in the management of chronic sinusitis. Indian J Otolaryngol Head Neck Surg 52:13–16, 1999. Sobol SE, Wright ED, and Frenkiel S. One-year outcome analysis of functional endoscopic sinus surgery for chronic sinusitis. J Otolaryngol 27:252–257, 1998. Stankiewicz J. Complications of endoscopic intranasal ethmoidectomy. Laryngoscope 97:1270, 1987. Stankiewicz J. Complications of endoscopic sinus surgery. Otolaryngol Clin North Am 22:749, 1989. Kennedy D, Shaman P, Deems D, et al. Complications of ethmoidectomy: A survey of Fellows of the American Academy of Otolaryngology–Head and Neck Surgery. Otolaryngol Head Neck Surg 111:589–599, 1994. Cumberworth VL, Sudderick RM, and Mackay IS. Major complications of functional endoscopic sinus surgery. Clin Otolaryngol 19:248–253, 1994. Roth M, and Kennedy D. Outcome and complications of surgical treatment. In Nasal Polpys: Epidemiology, Pathogenesis and Treatment. Settipane GA, Lund VJ, Bernstein JM, et al. (Eds). Providence, RI: Oceanside Publications, Inc., 177–185, 1997. Kurent Z, and Zargi M. Endoscopic polypectomy vs. endoscopic ethmoidectomy in combined treatment of massive bilateral nasal polyposis. Presented at the meeting of the E.R.S. and I.S.I.A.N., July 28–Aug. 1, 1998, Vienna, Austria. Bologna, Italy: Monduzzi Editore, 187–190, 1998. Venkatachalam VP, and Bhat A. Comparative evaluation of endoscopic and conventional surgical techniques in the management of nasal polyposis. JK Pract 5:295–299, 1998. Unlu HH, Caylan R, Nalca Y, et al. An endoscopic and tomographic evaluation of patients with sinusitis after endoscopic sinus surgery and Caldwell-Luc operation: A comparative study. J Otolaryngol 23:197–203, 1994. Levine HL. Functional endoscopic sinus surgery: Evaluation, surgery, and follow-up of 250 patients. Laryngoscope 100:79–84, 1990. Fawehinmi Y, Ardi T, and Dewan M. Nasal polyposis and fungal sinusitis–Efficacy of endoscopic sinus surgery. J Bahrain Med Soc 15(3):159–163, 2003. e

519

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.