1945892419851312
1945892419851312
1945892419851312
Abstract
Background: Corticosteroids are considered the most effective medication to control chronic rhinosinusitis with nasal
polyps (CRSwNP); however, the studies of systemic corticosteroids in this field are relatively few and meta-analyses are lacking.
Objective: We aimed to systematically review the efficacy and safety of systemic corticosteroids for patients with CRSwNP
via meta-analysis.
Methods: Data were extracted from relevant and appropriate randomized controlled trials (RCTs) of systemic corticosteroids
for patients with CRSwNP from PubMed, MEDLINE, Ovid, Embase, the Cochrane Central Register of Controlled Trials, and
Google Scholar. Efficacy was assessed based on clinical outcomes, while safety was assessed based on adverse events (AEs).
Results: A total of 337 relevant publications were identified, of which 7 RCTs including 414 participants were included in
the meta-analysis. Compared to placebos or nonsteroid treatments, systemic corticosteroids significantly improved nasal
obstruction scores standardized mean difference (SMD 2.81; 95% confidence interval [CI], 4.68 to 0.95; P ¼.003),
reduced the nasal polyp size (SMD 4.76; 95% CI, 6.99 to 2.52; P <.0001), and improved peak nasal inspiratory flow
(PNIF) (SMD 42.39; 95% CI, 28.95 to 55.84; P <.00001). The high-dose (more than or equal to 50 mg/day prednisone) and
low-dose subgroups (less than 50 mg/day prednisone) experienced similar benefits. However, insomnia and gastrointestinal
disturbances were noted more frequently in patients treated with high doses of prednisone. Other AEs were infrequent and
were not significantly different between the subgroups.
Conclusion: Systemic corticosteroids provide significant improvements in nasal symptoms and PNIF as well as a reduction
in nasal polyp size for patients with CRSwNP. Prednisone doses less than 50 mg/day were recommended when the efficacy of
oral corticosteroids in CRSwNP was balanced against potential adverse effects.
Keywords
chronic rhinosinusitis, meta-analysis, nasal polyps, randomized controlled trials, systemic corticosteroids
1
Department of Otolaryngology Head and Neck Surgery, Beijing TongRen
Hospital, Capital Medical University, Beijing, China
2
Beijing Key Laboratory of Nasal Diseases, Beijing Institute of
Introduction Otolaryngology, Beijing, China
3
Department of Allergy, Beijing TongRen Hospital, Capital Medical
Chronic rhinosinusitis with nasal polyps (CRSwNP) is a University, Beijing, China
chronic inflammatory disease of the nasal and paranasal Yunyun Zhang and Chengshuo Wang contributed equally to this work.
sinus mucosa for which the cause remains unclear.1
Corresponding Authors:
CRSwNP is often comorbid with asthma and other respi- Hongfei Lou, Department of Otolaryngology Head and Neck Surgery,
ratory diseases such as cystic fibrosis, primary ciliary dys- Beijing Tongren Hospital, Capital Medical University, No. 1,
kinesia, and aspirin intolerance triad.2–4 Nasal Dongjiaominxiang, Dongcheng District, Beijing 100730, China.
Email: louhongfei@yahoo.com
obstruction and loss of smell can be troublesome, limiting
Luo Zhang, Beijing Institute of Otolaryngology, No. 17, HouGouHuTong,
daily activities and the ability to sleep, thus reducing one’s DongCheng District, Beijing 100005, China.
well-being and quality of life (Qol).5 Nasal polyps can be Email: dr.luozhang@139.com
568 American Journal of Rhinology & Allergy 33(5)
Heterogeneity across study effects was assessed using due to pooled results from 2 independent studies
a v2 test and the I2 statistic. The I2 index describes the (n ¼ 2) and mixed CRS cohorts (polyp and nonpolyp
percentage of variation across studies due to heteroge- patients; n ¼ 2). Twelve RCTs met all inclusion criteria
neity rather than chance. A value of 0% indicates no for the systematic review (Table 1); however, 5 more
observed heterogeneity and larger values indicate RCTs were excluded for lacking quantifiable data.
increased heterogeneity. The values of P for hetero- Finally, 7 studies provided sufficient discrete data for a
geneity < .05 and/or I2 > 50% denoted significant het- formal meta-analysis.7,12–17
erogeneity. For instances of considerable heterogeneity, These 7 RCTs were conducted between 2001 and 2015
random-effects models were used instead of fixed-effects and comprised a total of 414 patients; sample sizes ranged
models. Furthermore, sensitivity analyses were also per- from 22 to 89 people and the trial durations spanned 7–14
formed to assess heterogeneity. days. All published articles were in English, and all par-
ticipants were over 18 years of age. The overall quality of
Results the RCTs was good with a low risk of bias (Figure 2).
Only articles that had adequate data available for pooled
Study Selection and Trial Characteristics analysis were included in the meta-analysis.
The search strategy identified 337 unique abstracts of
which 30 met the initial screening criteria (Figure 1).
Symptom Scores
An additional 14 studies were excluded due to a lack Nasal obstruction outcomes following oral corticoste-
of randomization (n ¼ 3), lack of placebo (n ¼ 3), or roids were measured in 5 trials; the 5 studies comprised
were only presented as abstracts (n ¼ 8). After reviewing a total of 280 participants. All studies demonstrated sub-
full-length articles, 4 additional studies were excluded jective nasal obstruction improvements, and the overall
1 Blomqvist et al.19 Europe Prednisolone Oral 260 10 Polyp score; nasal symptoms Nasal symptoms on the VAS (0–10) No
(Sweden) Polyp size on a scale of 0–3
2 Hissaria et al.8 Australia Prednisolone Oral 700 14 Polyp grade; nasal symptoms RSOM on a scale of 1–5 No
Polyp grade on a scale of 1–4
3 Alobid et al.14 Europe Prednisone Oral 270 14 Nasal symptoms; nasal polyp size Nasal symptoms on a scale of 0–3 Yes
(Spain) Polyp size on a scale of 0–3
4 Benitez et al.7 Europe Prednisone Oral 270 14 Nasal symptom; polyp size; Nasal symptoms on a scale of 0–3 Yes
(Spain) nasal patency Polyp size on a scale of 0–3
5 Kroflic et al.13 Europe Methylprednisolone Oral 420 7 Nasal symptom; polyp size Nasal symptoms on a scale of 0–3 Yes
(Slovenia) Polyp size on a scale of 0–3
6 Van Zele20 Europe Methylprednisolone Oral 320 21 Polyp grade; nasal symptoms Polyp grade on a scale of 0–4 No
(Belgium)
7 Vaidyanathan Europe Prednisolone Oral 350 14 Nasal polyp score; hyposmia; Polyp grade on a scale of 0–6 Yes
et al.17 (Scotland) total nasal symptom score; PNIF Hyposmia VAS (0–100)
8 Kirtsreesakul Asia Prednisolone Oral 700 14 Nasal symptoms; nasal polyp size Ranked symptom severity No
et al.18 (Thailand) on a scale of 0–6
9 Kirtsreesakul21 Asia Prednisolone Oral 700 14 Nasal symptoms; nasal polyp size Ranked symptom severity No
(Thailand) on a scale of 0–6
10 Alobid et al.15 Europe Prednisolone Oral 270 14 Nasal congestion; PNIF, Ranked symptom severity Yes
(Spain) CT scan score on a scale of 0–3
11 Alobid et al.16 Europe Prednisolone Oral 270 14 Nasal congestion; polyp size Ranked symptom severity Yes
(Spain) on a scale of 0–3
12 Ecevit et al.12 Europe Prednisolone Oral 720 14 Nasal symptom; polyp size; PNIF Nasal symptoms Yes
(Turkey) on the VAS (0–10)
Polyp size on a scale of 0–3, PNIF
Abbreviations: CT, computed tomography; PNIF, peak nasal inspiratory flow; RSOM, rhinosinusitis outcome measure; VAS, visual analog scale.
Zhang et al. 571
SMD was 2.81 (95% CI, 4.68 to 0.95) (Figure 3 following systemic corticosteroid use did not have orig-
(A)). Nasal obstruction improvements favored oral cor- inal data for meta-analysis.
ticosteroids over nonsteroid treatments (P ¼ .003). The Besides nasal obstruction, the loss of smell was anoth-
remaining 2 studies that assessed nasal obstructions er main complaint in nasal polyp patients. Olfactory out-
comes were measured by 7 studies in which 5
demonstrated significant improvements; the SMD of
the pooled studies was 1.93 (95% CI, 3.35 to
0.51) (Figure 3(B)). Olfactory outcome improvements
favored oral corticosteroids over nonsteroid treatments
(P ¼ .008). The other 2 studies showed statistically sig-
nificant improvements in olfactory outcomes after sys-
temic corticosteroid treatment compared to placebos
(P < .05); however, there were no sufficient original
data to include in the meta-analysis.
Figure 3. A pooled analysis of all RCTs using oral corticosteroids with reported nasal obstruction (A) and loss of smell (B) outcomes. A
negative SMD in this analysis favors the experimental group because the decrease in the nasal symptom scores signifies improvement. CI,
confidence interval; I2, the variation in the risk ratio attributable to study heterogeneity; RCT, randomized clinical trial; SD, standard
deviation; SMD, standardized mean difference.
572 American Journal of Rhinology & Allergy 33(5)
Figure 4. Quantitative data and a forest plot of the total nasal polyp scores of eligible studies comparing systemic corticosteroid use with
controls. The SMD was 4.76 (95% CI, 6.99 to 2.52). The negative SMD in this analysis favors the experimental group because the
decrease in the nasal polyp score indicates improvement. CI, confidence interval; I2, the variation in the risk ratio attributable to study
heterogeneity; SD, standard deviation; SMD, standardized mean difference.
Figure 5. Quantitative data and a forest plot of the total PNIF of eligible studies comparing systemic corticosteroids with controls. The
mean difference was 42.39 (95% CI, 28.95–55.84). CI, confidence interval; I2, the variation in the risk ratio attributable to study het-
erogeneity; PNIF, peak nasal inspiratory flow; SD, standard deviation.
Figure 6. Quantitative data and forest plots of the subgroup analysis of symptom outcomes based on the prednisone dose. (A) Nasal
obstruction data. In the high-dose subgroup, SMD ¼ 1.87 (95% CI, 2.91 to 0.83; P ¼.0004); in the low-dose subgroup, SMD ¼ 4.18
(95% CI, 7.27 to 1.08; P ¼.008). (B) Loss of smell data. In the high-dose subgroup, SMD ¼ 0.99 (95% CI, 1.89 to 0.09; P ¼.003). In
low-dose subgroup, SMD ¼ 2.85 (95% CI, 5.14 to 0.56; P ¼.01). SD, standard deviation; SMD, standardized mean difference; CI,
confidence interval.
Subgroup Analysis of High and Low Doses of We also assessed nasal polyp scores as the objective
Systemic Steroids outcome. Two studies with high doses and 4 studies with
low doses were recruited in the subgroup analysis of
One study with at least 50 mg/day prednisone (high
nasal polyp scores. There was a significant improvement
dose) and 3 studies with less than 50 mg/day (low
in the high-dose (SMD 0.50; 95% CI, 0.93 to 0.07;
dose) were examined in the subgroup analysis. There
P ¼ .02; Figure 7) and low-dose subgroups (SMD 1.16;
was a significant difference in favor of the high-dose
95% CI, 1.87 to 0.46; P ¼ .001; Figure 7) in the nasal
(SMD 1.87; 95% CI, 2.91 to 0.83; P ¼ .0004;
polyp scores, superior to placebos. Meanwhile, there was
Figure 6(A)) and low-dose subgroups (SMD 4.18;
95% CI, 7.27 to 1.08; P ¼ .008; Figure 6(A)), com- no significant difference between the 2 sub-
pared to placebos, with respect to nasal obstruction. groups (P ¼ .11).
There was no significant difference between the 2 sub-
groups (P ¼ .17). Discussion
We also evaluated the effect on olfactory function
among subgroups. There was a significant benefit in The management of nasal polyps can be challenging and
both subgroups (SMD 0.99; 95% CI, 1.89 to often requires long-term medical treatment to control
0.09; P ¼ .003 in the high-dose subgroup; nasal symptoms. Corticosteroids are frequently
SMD 2.85; 95% CI, 5.14 to 0.56; P ¼ .01 in the employed to treat CRS due to their potent anti-
low-dose subgroup) (Figure 6(B)). Similarly, there inflammatory effects. Intranasal corticosteroids are con-
was no significant difference between the 2 sub- sidered the first choice for nasal polyp treatment.
groups (P ¼ .14). Correspondingly, the EPOS 20126 also indicated that
574 American Journal of Rhinology & Allergy 33(5)
Figure 7. Quantitative data and a forest plot of the subgroup analysis of nasal polyp scores based on the prednisone dose. In the high-
dose subgroup, SMD ¼ 0.50 (95% CI, 0.93 to 0.07; P ¼.02). In the low-dose subgroup, SMD ¼ 1.16 (95% CI, 1.87 to 0.46;
P ¼.001). SD, standard deviation; SMD, standardized mean difference; CI, confidence interval.
short-term oral corticosteroids can be used to control that revealed an increase in AEs in the oral steroids
symptoms in patients with severe nasal polyps. Three group, compared to placebos, after a 50 mg/day predni-
studies7,15,16 using oral corticosteroids following intrana- sone treatment for 14 days for insomnia. Similarly, dys-
sal corticosteroids showed sustained benefits in nasal pepsia and gastrointestinal disturbances were noted
polyp management. A systematic review22 reported more frequently in patients with the same treatment.18
that more thorough, well-designed, prospective, RCTs However, no patients reported significant adverse symp-
are still needed to address oral steroid use for toms during the days immediately after prednisolone
CRSwNP. Our meta-analysis validates the efficacy of cessation. In our review, 50 mg/day prednisone for
systemic corticosteroid therapy in patients with nasal 14 days led to more frequent AEs in the experimental
polyps with subjective and objective clinical outcomes. group. The subgroup analysis revealed that a prednisone
These findings showed similar olfactory function dose of at least 50 mg/day did not offer better
improvements as the meta-analysis by Banglawala efficacy when compared a lower dose. Therefore, a pred-
et al.23Consistent with the other studies,10,19,24,25 we nisone dose of less than 50 mg/day is recommended,
found that systemic corticosteroids improved nasal considering the lower risk of AEs. Nevertheless, this rec-
obstructions and reduced polyp size. In this regard, ommendation should be tested with more studies in
Head et al.26 showed comparable data in a review with- the future.
out a quantitative meta-analysis. Poetker et al.27 evalu- A limitation of this analysis is that there was signifi-
ated steroid use and strongly recommended oral steroids cant variability in the duration (range: 7–21 days) and
for CRSwNP, which was consistent with our meta- total dose (range: 270–720 mg) of systemic corticoste-
analysis results. Moreover, our meta-analysis first roids in the intervention group. Thus, there was no uni-
demonstrated PNIF improvements after systemic corti- versal protocol of systemic corticosteroids for CRSwNP.
costeroid treatment, coinciding with advances in symp- Furthermore, we cannot define the correlation between
tom control and nasal polyp reduction. Also, the dose and efficacy. Another limitation is the lack of a
robustness of the overall effect of a single study provided long-term follow-up for systemic corticosteroid therapy,
strong evidence to support our results. negating a quantitative pooled analysis. Also, only 2
The benefits of systemic steroids are time limited due studies utilized a validated Qol instrument as the prima-
to the side effects.6 Therefore, the efficacy of oral corti- ry outcome measure, so we could not evaluate Qol via
costeroids in CRSwNP must be carefully balanced meta-analysis. Finally, instead of nasal obstruction and
against potential adverse effects. Vaidyanathan et al.17 loss of smell, other common symptoms of CRSwNP
reported safety outcomes, including basal and dynamic including facial pain and rhinorrhea, which did not
adrenal function, and observed markers of osteoblast have enough quantitative data in the 12 RCT studies,
activity. Adrenal function was suppressed by oral pred- were excluded from the meta-analysis. Interestingly,
nisolone but recovered after 10 weeks. No serious AEs, most of the studies in our review originated in Europe
such as osteonecrosis of the femoral head, were reported. (10/12); this could decrease the external validity and gen-
Hissaria et al.8 included an AEs analysis in their study eralizability to other patient populations.
Zhang et al. 575
18. Kirtsreesakul V, Wongsritrang K, Ruttanaphol S. Clinical 24. Stewart RA, Ram B, Hamilton G, Weiner J, Kane KJ.
efficacy of a short course of systemic steroids in nasal pol- Montelukast as an adjunct to oral and inhaled steroid ther-
yposis. Rhinology. 2011; 49(5):525–532. apy in chronic nasal polyposis. Otolaryngol Head Neck
19. Blomqvist EH, Lundblad L, Anggard A, Haraldsson PO, Surg. 2008;139(5):682–687.
Stj€arne P. A randomized controlled study evaluating med- 25. Wright ED, Agrawal S. Impact of perioperative
ical treatment versus surgical treatment in addition to med- systemic steroids on surgical outcomes in patients
ical treatment of nasal polyposis. J Allergy Clin Immunol with chronic rhinosinusitis with polyposis: evaluation
2001;107(2):224–228. with the novel Perioperative Sinus Endoscopy
20. Van Zele T, Gevaert P, Holtappels G, et al. Oral steroids (POSE) scoring system. Laryngoscope. 2007;117(11 Pt 2
and doxycycline: Two different approaches to treat nasal Suppl 115):1–28.
polyps. J Allergy Clin Immunol. 2010;125(5):1069–1076. 26. Head K, Chong LY, Hopkins C, Philpott C, Burton MJ,
21. Kirtsreesakul V, Wongsritrang K, Ruttanaphol S. Does Schilder AG. Short-course oral steroids alone for chronic
oral prednisolone increase the efficacy of subsequent rhinosinusitis. Cochrane Database Syst Rev
nasal steroids in treating nasal polyposis?. Am J Rhinol. 2016;4:CD011991.
2012;26(6):455–462. 27. Poetker DM, Jakubowski LA, Lal D, Hwang PH,
22. Martinez-Devesa P, Patiar S. Oral steroids for nasal Wright ED, Smith TL. Oral corticosteroids in the
polyps. Cochrane Database Syst Rev. 2011;(7):CD005232. management of adult chronic rhinosinusitis with and
23. Banglawala SM, Oyer SL, Lohia S, Psaltis AJ, Soler ZM, without nasal polyps: an evidence-based review with
Schlosser RJ. Olfactory outcomes in chronic rhinosinusitis recommendations. Int Forum Allergy Rhinol
with nasal polyposis after medical treatments: a systematic 2013;3(2):104–120.
review and meta-analysis. Int Forum Allergy Rhinol.
2014;4(12):986–994.