Keywords laryngotracheal stenosis - mitomycin C - systematic review - dilation - endoscopic
treatment
Introduction
The increasing number of patients with postintubation laryngotracheal stenosis has
been a challenge because of the lack of an effective treatment modality. The objective
of an effective laryngotracheal stenotic treatment should be to ensure adequate airway
diameter to allow ventilation and provide symptom relief, which are not achieved in
most patients. In many cases, endoscopic treatments and open surgical techniques are
used in the same patient; however, when surgical resection is not indicated, endoscopic
treatment options are adopted, which have an immediate effect.[1 ]
Topical agents have been used as adjuvants in endoscopic treatment, aiming to minimize
recurrence rates and prolong the symptom-free period after the intervention. The adjuvant
options studied in animal and human models include steroid injections, topical mitomycin
C (MMC), topical heparin, 5-fluorouracil, and halofuginone.[2 ]
Of these options, MMC is highlighted as an antibiotic and antineoplastic agent that
inhibits the proliferation of fibroblasts, thereby modulating the healing processes.
There is a long history of studies published regarding laryngology research on the
effect of MMC on laryngotracheal lesion models in dogs, rabbits, pigs, and rats. The
success rate in maintaining the dilation of the stenosis over a long period ranges
from 40% to 70%, with an average of 50%.[3 ] Satisfactory results have expanded the search for methods to improve the success
rate,[3 ] but the overall results of the use of MMC in humans are conflicting.
Although surgical resection of laryngotracheal stenosis with end-to-end anastomosis
is effective, conservative options are necessary to treat patients for whom surgery
is not indicated. Therefore, the present systematic review was conducted to evaluate
the efficacy of MMC as an adjuvant in the endoscopic treatment of laryngotracheal
stenosis.
Review of the Literature
Methodology
The present systematic review included studies published on the use of topical MMC
in the endoscopic treatment of laryngotracheal stenosis. The study design (review
of the literature) enabled the exemption of approval from the Research Ethics Committee.
Inclusion criteria: experimental or observational studies (with at least 5 participants
per group, with more than 80% of the patients being adolescents or adults, excluding
studies that only address children) evaluating endoscopic interventions with adjuvant
topical MMC being at least one of the treatment options for laryngotracheal stenosis.
Participants: patients with laryngotracheal stenosis. Interventions: endoscopic interventions
(mechanical dilation; use of electric scalpel, argon, harmonic, or laser; and cryotherapy)
in which MMC was administered to at least one participant group. Outcomes: the primary
outcomes include complete or partial resolution of laryngotracheal stenosis, characterized
by symptom-free period ≥1 year; the secondary outcomes include the number of procedures
with or without the administration of MMC (single or multiple procedures); and complications
characterized by the need for immediate intervention or other procedures (acute obstruction,
tracheostomy, dysphonia, subcutaneous emphysema, vocal fold laceration, temporary
vocal paralysis, among others).
We conducted a search for relevant literature on the following electronic databases:
PubMed (from 1966 to November 27, 2018); Embase (from 1973 to November 27, 2018);
LILACS (from 1982 to November 27, 2018); Cochrane (from 1993 to November 27, 2018);
Web of Science (from 1900 to November 27, 2018); and Clinical Trials (accessed on
November 27, 2018). The following search terms were used to find studies on the Medline
(PubMed) database: (tracheal stenosis or stenoses , tracheal or stenosis , tracheal or tracheal stenoses ) and (bronchoscopy or bronchoscopies or bronchoscopic surgical procedures or bronchoscopic surgical procedure or surgical procedure , bronchoscopic or surgical procedures , bronchoscopic or bronchoscopic surgery or bronchoscopic surgeries or surgeries , bronchoscopic or surgery , bronchoscopic ) and (placebos or sham treatment or mitomycin or mitomycin C or mitomycin-C or mitocin-C or mitocin C or mitocinC or NSC-26980 or NSC 26980 or NSC26980 or ametycine or mutamycin or adrenal cortex hormones or hormones , adrenal cortex or corticosteroids or corticoids or lasers or laser or Q-switched lasers or laser , Q-switched or lasers , Q-switched or Q-switched lasers or Q-switched laser or pulsed lasers or laser , pulsed or lasers , pulsed or pulsed laser or continuous wave lasers or continuous wave laser or laser , continuous wave or lasers , continuous wave or masers or maser or argon or cryotherapy or cryotherapies or therapy , cold or cold therapies or therapies , cold or cold therapy ). The same search strategy was applied to each database to achieve greater sensitivity.
There was no restriction regarding the language. The reference lists of relevant publications
found were analyzed to search for studies not identified with the search strategy.
Selection of studies: two researchers (TLOQ and DCC) independently examined the titles
and abstracts to eliminate irrelevant articles, retrieved relevant full texts, identified
the location and environment of the studies, the details of the intervention, the
date and duration of the studies, and assessed the full texts for compliance with
the eligibility criteria. Study authors were contacted, if necessary, to clarify questions
related to the study and to decide on its inclusion in this review. Any disagreements
were resolved by consensus and discussion with the participation of all of the authors.
Data extraction and management: data from the selected studies were extracted independently
by the two reviewers (TLOQ and DCC). A standard form was used to obtain the following
information from the extracted studies: type of treatment, total number of participants,
number of patients for each intervention, follow-up period, number of interventions
performed, evaluated outcomes, type of study, age of the patients and participants,
and initial and final classification of laryngotracheal stenosis.
Assessments of the risk of bias were performed for randomized controlled trials (RCTs)
by using the “Risk of Bias” tool for Cochrane reviews. These assessments were not
performed for observational studies because of the lack of consensus for their application
in these studies; however, we consider that they are biased and subject to the effect
of confounding factors. The two reviewers independently evaluated each study. Any
disagreements were resolved by consensus, or by consultation with the third author
(RHGM).
Measures of treatment effect: two measurement forms for the interventional effect
were used: a proportional meta-analysis using the StatsDirect (StatsDirect Ltd, Birkenhead,
Merseyside, UK) software, version 3.0.121, was performed for the studies that only
evaluated the effects of the MMC intervention.[4 ] Forest plots were drawn to summarize dichotomous outcome data presented as a proportion
with their corresponding 95% confidence intervals (95%CIs). The Review Manager (RevMan,
The Nordic Cochrane Centre, Copenhagen, Denmark) software, version 5.3 was used to
analyze studies that compared the effects of interventions with and without MMC. The
dichotomous outcomes analyzed were considered with a 95%CI. The results were also
summarized using a forest plot.
To quantify the inconsistencies of the studies used in the meta-analysis, we conducted
the heterogeneity test I2 = [(Q − df)/Q] × 100%, in which Q is the Chi squared, and df is the degree of freedom.
In the presence of heterogeneity between the studies, we used the random effect model
of the meta-analysis, leaving the fixed-effect model only for heterogeneities equal
to zero. The random effect model was planned to perform subgroup analyses according
to the location and severity of the stenosis.
Results
Search results: an initial search for relevant studies was conducted in October 2015,
and another search was conducted on November 27, 2018. With these searches, we identified
a total of 723 articles from PubMed, Embase, LILACS, Cochrane, and Web of Science
databases, and 1 ongoing study in the Clinical Trials database.[18 ] [Fig. 1 ] is the flowchart of the selection process of all studies.
Fig. 1 Flowchart showing the studies identified and evaluated during the review.
Included studies: in total, 15 studies[1 ]
[3 ]
[5 ]
[6 ]
[7 ]
[8 ]
[9 ]
[10 ]
[11 ]
[12 ]
[13 ]
[14 ]
[15 ]
[16 ]
[17 ] (written in English: 14; written in Portuguese: 1) published from 2001 to 2018 (case
series: 12; prospective, randomized: 1; case-control: 1; prospective cohort: 1) were
included in the present review ([Table 1 ]). Of these studies, 10 were conducted in the United States, one in Portugal, one
in the United Kingdom, one in India, and one in Brazil, and the last one was a multicenter
(Thailand and Germany) study ([Fig. 1 ]).
Table 1
Characteristics of the included studies
Author (year) and country
Type of study (period studied)
Total number of patients (type of stenosis)
Concentration of MMC (time of application)
Intervention performed
Follow-up
Primary outcome: symptom-free time ≥1 year/total
Secondary outcomes
Number of procedures required
Complications
Rahbar et al[5 ] (2001), United States
Case series (January 1998–December 1999)
15 (10 subglottic, 3 glottic, 2 tracheal)
0.4 mg/mL (4 minutes)
Radial incision with CO2 laser + dilation + MMC
6–29 months; < 1 year = 1 pat; 1–2 years = 10 pat; > 2 years = 4 pat
11/15
1 proc = 9 pat; > 1 proc = 6 pat (2 = 5; 3 = 1)
Tracheostomies (3)
Poor quality voice (3)
Perepelitsyn and Shapshay[6 ] (2004), United States
Case-control (1990–1997 without MMC; 1997–2001 with MMC)
20 without MMC (3 glottic, 17 subglottic); 16 with MMC (8 glottic, 8 subglottic)
0.4 mg/mL (4 minutes)
Only CO2 laser or laser with MMC or laser with corticoid
Without MMC = 25.2 months; with MMC = 5 to 70 months
Without MMC: 5/20; with MMC: 12/16
2.35 per patient
Zero
Nouraei et al[7 ] (2006), United Kingdom
Case series (2003–2005)
31, but 11 were recent, (20 late stenoses)
1 mg/mL (3 minutes)
Radial incision with CO2 laser + dilation + MMC
15.4 ± 1.8 months
14/20
1 proc = 9 pat; > 1 proc = 11 pat
not reported
Schweinfurth[8 ] (2006), United States
Case series (does not mention period)
20 (degree 1 = 1; degree 2 = 3; degree 3 = 10; degree 4 = 6)
1 mg/mL (5 minutes)
Incision with CO2 laser + dilation + MMC
Mean of 16 months (5 to 27 months)
15/20; tracheostomy (2), tracheal resection (3), death (1)
1 proc = 15 pat; > 1 proc = 5 pat (2 = 4; 3 = 1)
Tracheostomy (2) Dysphonia (4)
Simpson and James[9 ] (2006) United States
Case-control (January 1999–December 2004)
36 (24 subglottic, 5 glottic, 3 supraglottic, 4 combined); 29 with MMC and 7 without
MMC
2 groups; 0.4 mg/mL: 22 patients; 1 mg/mL: 7 patients (5 minutes)
CO2 laser + dilation or CO2 laser + dilation + MMC
37.4 months
Mean symptom-free time without MMC: 4.9 months; with MMC: 23.2 months
With MMC: 1 proc = 24 pat; > 1 proc = 5 pat (2 = 4; 3 = 1); Without MMC: 2 proc = 2
patients; ≥3 proc = 5 patients
not reported
Ubell et al[10 ] (2006), United States
Case series (August 2001–August 2005)
50 (3 supraglottic, 13 glottic, 15 subglottic, 5 tracheal, 14 combined)
0.4 mg/mL (Does not mention time of application)
CO2 laser + dilation + MMC
Not reported
Not reported
1.86/patient
Fungal infection (1)
Roediger et al[11 ] (2008), United States
Case series (October 2004–January 2008)
15 subglottic
0.5 mg/mL (3 minutes)
CO2 laser + dilation + MMC
Mean of 18 months (3 to 39 months)
10/15
1 proc = 6 pat; > 1 proc = 9 pat (2 = 6; 3 = 2; 4 = 1)
Acute Lumen Obstruction (1)
1 proc = 12 pat;
2 proc = 14 pat
1 proc = 4 pat;
2 proc = 3 pat
With MMC: 1 proc = 3 pat;
> 1 proc = 7 pat (2 =2; 3 = 5);
without MMC: > 1 proc = 2 pat
(5 = 1; 7 = 1)
> 2 proc = 29 pat;
≤2 proc= 51 pat;
220 procedures;
interval between procedures: 405 days;
more distal stenosis = more frequent procedures
Smith and Elstad[3 ] (2009), United States
Case series (does not mention period)
26 subglottic
0.5 mg/mL (5 minutes)
1st - CO2 laser + dilation + MMC; 2nd - CO2 laser + dilation + (MMC or placebo)
5 years
21/26
1 proc = 6 pat;
> 1 proc = 9 pat (2 = 6; 3 = 2; 4 = 1)
Zero
1 proc = 12 pat;
2 proc = 14 pat
1 proc = 4 pat;
2 proc = 3 pat
With MMC: 1 proc = 3 pat;
> 1 proc = 7 pat (2 =2; 3 = 5);
without MMC: > 1 proc = 2 pat
(5 = 1; 7 = 1)
> 2 proc = 29 pat;
≤2 proc= 51 pat;
220 procedures;
interval between procedures: 405 days;
more distal stenosis = more frequent procedures
Madan et al[12 ] (2012), India
Case series (July 2009–August 2011)
7 (degree III)
0.4 mg/mL (2 minutes in each quadrant)
Dilation + MMC
27 days
1/7
1 proc = 6 pat;
> 1 proc = 9 pat
(2 = 6; 3 = 2; 4 = 1)
not reported
1 proc = 12 pat;
2 proc = 14 pat
1 proc = 4 pat;
2 proc = 3 pat
With MMC: 1 proc = 3 pat;
> 1 proc = 7 pat (2 =2; 3 = 5);
without MMC: > 1 proc = 2 pat
(5 = 1; 7 = 1)
> 2 proc = 29 pat;
≤2 proc= 51 pat;
220 procedures;
interval between procedures: 405 days;
more distal stenosis = more frequent procedures
Gouveris et al[13 ] (2013), Germany and Thailand
Case-control (does not mention period)
12 subglottic (7 degree II and 5 degree I); 10 with MMC and2 without MMC
5 mg (10 minutes)
CO2 laser + MMC or only CO2 laser
With MMC: 7 to 72 months; without MMC: 18 to 115 months
8/10 with MMC; 2/2 without MMC
1 proc = 6 pat;
> 1 proc = 9 pat
(2 = 6; 3 = 2; 4 = 1)
not reported
1 proc = 12 pat;
2 proc = 14 pat
1 proc = 4 pat;
2 proc = 3 pat
With MMC: 1 proc = 3 pat;
> 1 proc = 7 pat (2 =2; 3 = 5);
without MMC: > 1 proc = 2 pat
(5 = 1; 7 = 1)
> 2 proc = 29 pat;
≤2 proc= 51 pat;
220 procedures;
interval between procedures: 405 days;
more distal stenosis = more frequent procedures
Parker et al[1 ] (2013), United States
Case series (March 2000–December 2010)
80 (does not mention the type)
1 mg/mL (4 minutes)
Cold incision + balloon dilation + MMC application + steroid injection; open surgery = 8
2.9 years
Not reported
1 proc = 6 pat;
> 1 proc = 9 pat
(2 = 6; 3 = 2; 4 = 1)
Dyspnea (2) Subcutaneous emphysema (1) Laceration of vocal fold (1) Temporal vocal
fold paralysis (1)
1 proc = 12 pat;
2 proc = 14 pat
1 proc = 4 pat;
2 proc = 3 pat
With MMC: 1 proc = 3 pat;
> 1 proc = 7 pat (2 =2; 3 = 5);
without MMC: > 1 proc = 2 pat
(5 = 1; 7 = 1)
> 2 proc = 29 pat;
≤2 proc= 51 pat;
220 procedures;
interval between procedures: 405 days;
more distal stenosis = more frequent procedures
Viveiros et al[14 ] (2013), Portugal
Case series (2006–2012)
11 (9 complex, 2 simple)
0.4 mg/ml (3 minutes)
Dilation with rigid bronchoscope + MMC
27.5 months
6/11
1 proc = 5 pat;
> 1 proc = 6 pat
(2 = 2; > 2 = 4)
not reported
1 proc = 6 pat;
> 1 proc = 5 pat
(2 = 4; 3 = 1)
With MMC: 2,3/patients;
without MMC: 2,0/patients;
interval between procedures:
375 days with MMC;
186 days without MMC
1 proc = 5 pat;
> 1 proc = 17 pat
(2 = 8; 3 = 3; 4 = 4; 5 = 2)
Vorasubin et al[15 ] (2014), United States
Case series (2008–2011)
11 (6 subglottic, 2 tracheal, 3 combined)
0.4 mg/mL (4 minutes)
CO2 laser + dilation + MMC
28 months
9/11
1 proc = 5 pat; > 1 proc = 6 pat (2 = 2; > 2 = 4)
not reported
1 proc = 6 pat;
> 1 proc = 5 pat
(2 = 4; 3 = 1)
With MMC: 2,3/patients;
without MMC: 2,0/patients;
interval between procedures:
375 days with MMC;
186 days without MMC
1 proc = 5 pat;
> 1 proc = 17 pat
(2 = 8; 3 = 3; 4 = 4; 5 = 2)
Reichert et al[16 ] (2015), United States
Case-control (January 2005–May 2013)
37 subglottic; 15 without MMC and 22 with MMC
0.4 mg/mL (4 minutes)
CO2 laser or CO2 laser + MMC
10 months
15/22 with MMC; 6/15 without MMC
1 proc = 5 pat; > 1 proc = 6 pat (2 = 2; > 2 = 4)
Zero
1 proc = 6 pat; > 1 proc = 5 pat (2 = 4; 3 = 1)
With MMC: 2,3/patients; without MMC: 2,0/patients; interval between procedures: 375
days with MMC; 186 days without MMC
1 proc = 5 pat; > 1 proc = 17 pat (2 = 8; 3 = 3; 4 = 4; 5 = 2)
Cataneo et al[17 ] (2018), Brazil
Prospective cohort (2003–2010)
22 (4 subglotic, 14 tracheal, 4 combined)
0.5 mg/mL (2 minutes)
Dilation + MMC
> 1 year
14/22
1 proc = 5 pat; > 1 proc = 6 pat (2 = 2; > 2 = 4)
Mediastinal emphysema (2) Granuloma (1)
1 proc = 6 pat;
> 1 proc = 5 pat
(2 = 4; 3 = 1)
With MMC: 2,3/patients;
without MMC: 2,0/patients;
interval between procedures:
375 days with MMC;
186 days without MMC
1 proc = 5 pat;
> 1 proc = 17 pat
(2 = 8; 3 = 3; 4 = 4; 5 = 2)
Abbreviations: CO2, carbon dioxide; MMC, mitomycin C; pat, patient; proc, procedure;
RCT, randomized clinical trial.
Characteristics of the included studies: the selected studies involved a total of
387 patients aged between 2 and 79 years (mean age: 46 years). The gender of the patients
was only mentioned in 12 studies (men: 116; women: 199). In the remaining 3 studies
(72 patients), the gender was not mentioned. Laryngotracheal stenosis was diagnosed
by anamnesis, physical examination, imaging methods, endoscopic examinations and,
in some cases, intraoperatively. The classification of stenosis in each study differed,
and not all studies classified the degree of stenosis at the time of diagnosis before
the interventions. Of the 387 patients, 29 had diagnoses of glottic stenosis, 135
of subglottic stenosis, 7 of supraglottic stenosis, 23 of tracheal stenosis, and 86
of combined-site stenosis. The remaining 107 patients were distributed among 6 studies
in which the location of the stenosis was not mentioned. Of the 387 patients, the
degree of stenosis was only determined in 64 patients. It was not possible to perform
subgroup analyses because the outcomes were not separated by location and severity
of the stenosis. Two different interventions, with and without MMC, were performed
in four studies.[6 ]
[9 ]
[13 ]
[16 ] In the remaining studies, the intervention was always with MMC. The type of intervention
in each study, the dose and time of MMC application, the number of procedures performed,
the follow-up period, and the treatment outcomes are detailed in [Table 1 ].
Risk of bias: only one study was an RCT in which MMC was used in all patients in the
first procedure, and all of them were randomized to receive either MMC or placebo
in the second procedure.[3 ] This randomization was performed to test whether two applications of MMC were better
than one application. As MMC was administered to all patients once or twice, they
were used in the meta-analysis, so despite the fact that it was an RCT study, the
data were used as a series of cases. The other studies had an observational design
and, therefore, were subject to the effect of confounding factors. A prospective,
double-blinded, randomized, placebo-controlled ongoing study was identified in the
Clinical Trials database (NCT01523275). The University of California, San Francisco,
has been recruiting participants since January 2012, the study is scheduled to be
completed by January 2019, and no preliminary data had been published until November
27, 2019. This study includes individuals older than 18 years with laryngotracheal
stenosis, and it aims to determine if the application of MMC associated with radial
incision using CO2 laser and dilation increases the interval between the endoscopic surgical procedures,
improves the symptom-free period, and determines the maximum inspiratory flow in patients
with laryngotracheal stenosis at a follow-up period of 24 months.[18 ]
Effects of the Interventions (Meta-Analysis)
Full or partial resolution: of the 15 studies selected, 12 analyzed the symptom-free
period after the use of MMC (n = 195). The rate of symptom-free periods ≥1 year was of 69% with the use of MMC (95%CI:
61–76%, I2 = 17.3%) ([Fig. 2 ]).
Fig. 2 Forest plot of symptom-free period ≥1 year for patients receiving MMC. The proportion
of patients with this outcome was of 0.69 (95%CI: 0.61–0.76).
Four studies analyzed the symptom-free period with and without the use of MMC. However,
only 3 of these studies (n = 85, 48 with and 37 without MMC) reported the number of patients who remained symptom
free for ≥1 year, which was 73% for MMC users and 35% for non-MMC users. The other
study presented the mean symptom-free period of 23.2 months for those who received
MMC, and of 4.9 months for those who did not receive MMC. The meta-analysis comparing
these two groups showed that the odds of being symptom-free for ≥1 year was 4 times
higher in MMC-treated patients (OR: 4.23; 95%CI: 1.47–12.21; I2 = 11%) ([Fig. 3 ]).
Fig. 3 Forest plot of symptom-free period ≥1 year. The chance for this outcome was greater
in patients who received MMC (OR: 4.23; 95%CI: 1.47–12.21; I2 = 11%).
Number of procedures required: in 11 of the studies, it was possible to separate the
patients into 2 groups, those submitted to a single procedure and those who underwent
more than one procedure. A single procedure was performed in 52% of the patients (95%CI:
39–64%; I2 = 64.7%) ([Fig. 4A ]) and more than 1 procedure was performed in 48% of the patients (95%CI: 36–61%;
I2 = 64.7%) ([Fig. 4B ]).
Fig. 4 (A ) Proportion of patients who underwent a single procedure (0.52; 95%CI: 0.39–0.64).
(B ) Proportion of patients who underwent more than one procedure (0.48; 95%CI: 0.36–0.61).
Complications: The complications were considered to be related to the entire endoscopic
procedure, and not only to the application of MMC. Of the studies analyzed, 6 reported
complications, and 3 reported no complications (complications: 9%; 95%CI: 3–18%; I2 = 79.8%; studies: 9; participants: 285) ([Fig. 5 ]). Six studies did not report this outcome. The complications included tracheostomy,
dysphonia, mediastinal and subcutaneous emphysema, fungal infection at the stenotic
site, acute light obstruction, restenosis, vocal fold laceration, granuloma, and temporary
vocal paralysis.
Fig. 5 Proportion of complications in 9 studies (0.09; 95%CI: 0.03–0.18).
Other outcomes assessed: one study evaluated the total cost involved in treating patients
with MMC.[10 ] Endoscopic treatment using MMC was reported to cost US$455 per patient, and open
surgery, US$7,840 (17 times more expensive). The study authors concluded that the
endoscopic treatment is economical if only 1 in 17 patients does not require open
surgery. One study measured airway resistance (AR) before and after the procedure.[13 ] The average AR was reduced by almost half, from 1.004 kPa/L/s before the procedure
to 0.526 kPa/L/s after the procedure.
Discussion
The surgical treatment of laryngotracheal stenosis by resection of the stenotic segment
followed by end-to-end anastomosis has been shown to be highly effective; however,
not all patients are indicated for this surgery. The endoscopic treatment that offers
dilation of the stenotic area is an alternative option for such patients, although
scarring of the bloody area during the treatment leads to recurrence of the stenosis.
Drugs that prevent the proliferation of fibroblasts, such as MMC, are reported to
delay restenosis. Mitomycin C was first used for airway dilation in the 1970s, and
since then, the results of the studies have been ambiguous, with some proving its
efficacy[19 ]
[20 ] and others attesting its ineffectiveness.[12 ]
Despite the fact that the prevalence of laryngotracheal stenosis is on the rise, the
incidence is relatively low. This rise can be attributed to treatment success in intensive
care units, which are currently saving several patients who in the past would have
relatively low survival rates. Perhaps because of the low incidence, there are no
published randomized studies comparing the use of MMC with placebo or other drugs,
and healthcare providers have to rely on observational studies for decision making,
and even carry out systematic reviews with them, until randomized studies are performed.[21 ]
The present review indicates that MMC is effective in the treatment of laryngotracheal
stenosis, with a resolution probability characterized by symptom-free period ≥1 year
of ∼ 70%. This result is promising and optimistic for the adjuvant use of topical
MMC in the conservative treatment of laryngotracheal stenosis of varying degrees and
durations, as the studies evaluated patients with various types of stenosis with different
characteristics. Most of the studies in the present review showed that the benefits
outweighed the risks associated with the use of MMC, and that the symptom-free period
was ≥1 year in most patients.
The meta-analysis of the studies comparing endoscopic treatments with and without
MMC showed that the chance of achieving a symptom-free period ≥1 year is 4 times higher
in patients receiving MMC. However, the number of studies evaluating patients who
did not receive MMC was considerably low, with a small sample size, resulting in a
large dilation of the CI. Therefore, the evidence will have to come from future studies.
Regarding the number of procedures required, we found no difference between the patients
who underwent a single procedure and those who were submitted to more than one procedure.
Moreover, we found only 1 randomized, prospective, double-blinded, placebo-controlled
study with 26 patients with laryngotracheal stenosis. In the first procedure, all
patients underwent a radial laser incision using CO2 laser, dilation, and topical application of MMC (0.5 mg/mL). In the second procedure,
after 1 month, these patients were randomized to receive either MMC or a placebo.
During the first 3 years of follow-up, the study showed that patients who received
MMC in both procedures had a higher rate of resolution of symptoms than those who
received MMC only in the first procedure. Nevertheless, after the fourth year, the
frequency of restenosis was almost the same between the two groups.
Complications were only reported in 9 studies, but with quite different incidences:
6 of them with low and 3 with a high frequency of complications, so the heterogeneity
among studies exceeded 75%, which makes the combination of studies inappropriate,
so the incidence of complications should be viewed with caution. These complications
were not specific to the procedure; they were also related to the disease, such as
fungal infection and the need for tracheostomy. The complications resulting from the
intervention were also not specific because of the use of MMC, but of the entire endoscopic
procedure, such as emphysema, dysphonia, laceration, or vocal fold paralysis and acute
light obstruction, which occurred with a low incidence. This incidence was calculated
based on the number of patients, and it was lower regarding the number of procedures,
because many of the patients underwent more than one procedure.
Most of the included studies were considered to be affected by confounding factors
because of their observational design, thereby reducing the quality of the evidence.
The only randomized study considered to have low risk of bias compared one with two
applications of MMC, not MMC with placebo; therefore, it cannot be used to increase
the quality of evidence in the present review.
At the time of the database search, we found a prospective, randomized, double-blinded,
placebo-controlled study that is still in the recruitment phase, which plans to use
topical MMC as adjuvant therapy to control stenosis. Whether this study presents a
relevant method and more reliable results following the application of MMC will only
be known after its publication.[18 ]
We are confident that well-focused clinical issues have facilitated the construction
of a comprehensive search strategy suitable to capture most of the relevant literature
published, minimizing the likelihood of missing relevant studies. We believe that
the well-defined criteria for the inclusion of the studies and the standardized and
unbiased extraction of data internally validated the review, although most of the
studies were observational.
The inclusion of observational studies in systematic reviews has already been done
in a Cochrane review.[22 ] Some characteristics of observational studies may reduce the risk of bias. For instance,
a low selection bias is introduced if the control group is selected from the same
population of cases; the performance bias is often high when the participants and
the professionals are not masked; and the detection bias may be low if the results
are obtained from medical records, resulting in the outcome assessors being independent.[21 ]
The broad search strategy applied to the various databases may have reduced the loss
of primary studies that used MMC in the conservative treatment of laryngotracheal
stenosis.
Despite the poor quality of the evidence, the present meta-analysis can be consulted
for decision-making on the use of topical MMC in the treatment laryngotracheal stenosis
as an alternative option. It also suggests the need for prospective studies comparing
MMC with placebo or other drugs to clarify the efficacy of the MMC therapy.
Final Comments
Evidence supports the effective and safe use of topical MMC as an adjuvant in the
endoscopic treatment of laryngotracheal stenosis.
Implications for the practice: topical MMC seems to provide a reasonable effectiveness,
with a symptom-free period ≥1 year. The safety in the performance of the procedure
is difficult to measure by the incidence of complications. This is due to the heterogeneity
among the studies, some with a high and others with a low incidence of complications.
Implications for research: there is a greater need for high-quality prospective, controlled
trials comparing the topical use of MMC with other types of conservative treatment
for laryngotracheal stenosis to provide an alternative treatment option for patients
not indicated for open surgery.
