Summary of findings 1. Intranasal steroid spray compared to no intervention/placebo for the prevention of persistent post‐COVID‐19 olfactory dysfunction.
| Intranasal steroid spray compared to no intervention/placebo for the prevention of persistent post‐COVID‐19 olfactory dysfunction | ||||||
| Patient or population: people with olfactory dysfunction for less than 4 weeks following COVID‐19 infection Setting: hospitalised or in isolation at home; studies conducted in Egypt, Iran and Turkey Intervention: intranasal steroid spray Comparison: no intervention or placebo | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with no intervention/placebo | Risk with intranasal steroid spray | |||||
| Presence of normal olfactory function Assessed by participants as a score of 10 on a VAS (range 0 to 10) ≤ 4 weeks |
Study population | RR 1.19 (0.85 to 1.68) | 100 (1 RCT) | ⊕⊝⊝⊝ very low1,2 | — | |
| 520 per 1000 | 619 per 1000 (442 to 874) | |||||
| Presence of normal olfactory function Assessed with psychophysical testing (Iran‐Smell Identification Test (Iran‐SIT), score ≥ 19/24) ≤ 4 weeks |
Study population | RR 2.31 (1.16 to 4.64) | 77 (1 RCT) | ⊕⊝⊝⊝ very low3,4 | — | |
| 211 per 1000 | 486 per 1000 (244 to 977) | |||||
| Serious adverse events | One study reported that no adverse events were identified during the study | Not estimable | 77 (1 RCT) |
⊕⊝⊝⊝ very low3,5 | — | |
| Change in sense of smell Assessed by participants (VAS, higher score = better, range 0 to 10) ≤ 4 weeks |
The mean change in sense of smell was 5.7 points | MD 0.5 points lower (1.38 lower to 0.38 higher) | — | 77 (1 RCT) | ⊕⊝⊝⊝ very low3,6 | No minimally important difference has been reported. We considered that a difference of 0.5 points was unlikely to be important to participants. |
| Change in sense of smell Assessed by participants (VAS, higher score = better, scale 0 to 10) > 4 weeks to 3 months |
The mean score for sense of smell was 6.1 points at 30 days of follow‐up | MD 2.4 points higher (1.32 higher to 3.48 higher) | — | 100 (1 RCT) | ⊕⊝⊝⊝ very low7,8 | No minimally important difference has been reported. We considered that a difference of 2.4 points may be important to participants. |
| Change in sense of smell Assessed with psychophysical testing (Iran‐SIT, higher = better, scale 0 to 24) ≤ 4 weeks |
The mean change in sense of smell was 7.9 points | MD 0.2 points higher (2.06 lower to 2.46 higher) | — | 77 (1 RCT) | ⊕⊕⊝⊝ low3,8 | No minimally important difference for the Iran‐SIT has been reported. We considered that a difference of 0.2 points was unlikely to be important to participants. |
| Prevalence of parosmia | This was not assessed or reported by any of the included studies. | |||||
| Change in sense of taste | This was not assessed or reported by any of the included studies. | |||||
| Disease‐related quality of life | This was not assessed or reported by any of the included studies. | |||||
| Other adverse effects | One study reported that no adverse events were identified during the study. | Not estimable | 77 (1 RCT) |
⊕⊝⊝⊝ very low3,5 | — | |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; RCT: randomised controlled trial; RR: risk ratio; VAS: visual analogue scale | ||||||
| GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect. | ||||||
1Serious risk of bias due to lack of blinding of participants, personnel or outcome assessors.
2Very serious imprecision as sample size smaller than optimal information size (take as 400 participants, as a rule of thumb) and confidence interval includes both potential harm and considerable benefit.
3Serious risk of bias due to unclear randomisation, blinding of outcome assessors, potential for selective reporting and other biases.
4Very serious imprecision as sample size is smaller than the optimal information size (taken as 400 participants, as a rule of thumb) and the confidence interval for the effect includes the potential for substantial benefit (766 more people per 1000) and a trivial benefit (36 more people per 1000)
5Very serious imprecision as sample size is smaller than the optimal information size (taken as 400 participants, as a rule of thumb) and an effect size cannot be calculated.
6Very serious imprecision as sample size is smaller than the optimal information size (taken as 400 participants, as a rule of thumb) and the confidence interval for the effect includes the potential for considerable harm from the intervention (up to 1.38 points lower) as well as a trivial benefit (up to 0.38 points higher).