The Prisma flowchart for study selection is presented in Fig. 1. Our initial search found 793 studies, of which 40 studies were included by screening their titles and abstracts. After screening the remaining studies full-text, six trials evaluating seven ultrasound-related outcomes were included in the final quantitative meta-analysis [11–16].
Table 1 summarizes the characteristic of included studies. The sample number of selected studies ranged from 30 to 93, and the total number of participants in all studies was 332. The number of women whose results were pooled for each outcome was 147 for MCA RI, 264 for MCA PI, 188 for MCA S/D ratio, 197 for UA RI, 284 for UA PI, 238 for UA S/D ratio, and 84 for CPR. The gestational age (GA) of women was 24 weeks and more.
Table 1
References | country | Sample size (N of loss to F/U) | Gestational age | Intervention | comparison | outcomes |
Dastjerdi MV et al (2012) (17) | Iran | 59 (18) | 24–37 weeks | tablet of Sildenafil citrate (50 mg) | placebo tablets | Changes in PI, RI and S/D of MCA and UMA, 2 hour following treatment |
El-Sayed MA et al (2017) (18) | Egypt | 54 (0) | 24 weeks or more | tablet of Sildenafil citrate (50 mg) | placebo tablets containing starch | Changes in PI, RI and S/D of MCA, UMA and UTA, and CPR 2 hour following treatment |
khan MI et al (2017) (19) | India | 93 (0) | 28 weeks or more | tablet of Sildenafil (25mg TID for a week and then increased to 50mg TID till delivery). | placebo tablets | Changes in PI, RI and S/D of MCA and UMA, 1 week following initiation of treatment |
Maged M et al (2018) (20) | Egypt | 50 (0) | 24–32 weeks | Tablet of Sildenafil (20 mg daily and then, if no side effects occurred, the dose increased to 20 mg TID till delivery). | N/A | Changes in PI, RI and S/D of UMA, 4 week following initiation of treatment |
Mohammad EE et al (2017) (21) | Egypt | 30 (0) | 26–32 weeks | Tablet of Sildenafil (20 mg daily for 6 weeks) | placebo tablets | Changes in PI and S/D of MCA and UMA, 4–6 week following initiation of treatment |
Shehata NA et al (2020) (22) | Egypt | 46 | 24–34 weeks | Tablet of Sildenafil (20 mg daily TID + fish oil + zinc supplementation | placebo tablets + fish oil + zinc supplementation | Changes in PI of MCA and UMA, 2 hour and 2 weeks following initiation of treatment |
Abbreviations. CPR, cerebroplacental ratio; MCA, middle cerebral artery; mg, milligram; N/A, not applicable; TID, Three times a day; RI, resistive index; S/D, systolic to the diastolic ratio; UA, umbilical artery. |
The main intervention component in all selected studies was the oral tablet of sildenafil citrate, given to women in a dose range between 50 to 75 mg daily until delivery [11–16]. In one study, supplementation with fish oil and zinc has been added to the treatment and comparison group [13]. Regarding the comparison group, in five studies, women received a placebo [11–15], and only in one study was nothing given to women [16].
The assessment of the risk of bias in the selected studies is presented in Table 2. Each study bias has been judged as low risk, unclear, and high risk of bias. Overall, 42 responses were recorded, of which 69.04 % was a low risk of bias presenting in all seven domains of the Cochrane tool. The status of four types of biases, including those related to random sequence generation, allocation concealment, blinding of participants and personnel, and blinding of outcome assessors, was unclear in 28.57 % of responses. The risk of attrition bias was high in only one study, which had a loss to a follow-up rate of 30 % [15]. The funnel plot was almost symmetric, showing no publication bias (Fig. 2). The statistical evaluation of publication bias through Egger’s regression and Begg's rank test was not-significant (p-value = 0.78 and p-value = 0.99, respectively).
Table 2
Quality assessment of selected studies
References | Random sequence generation (selection bias) | Allocation concealment (selection bias) | Blinding of participants and personnel (performance bias) | Blinding to outcome assessment (detection bias) | Incomplete outcome data (attrition bias) | Selective reporting (reporting bias) | Other bias |
Dastjerdi MV et al (2012) (17) | Unclear | Unclear | Low risk | Low risk | High risk | Low risk | Low risk |
El-Sayed MA et al (2017) (18) | Low risk | Low risk | Low risk | Low risk | Low risk | Low risk | Low risk |
khan MI et al (2017) (19) | Unclear | Unclear | Low risk | Low risk | Low risk | Low risk | Low risk |
Maged M et al (2018) (20) | Unclear | Unclear | Unclear | Unclear | Low risk | Low risk | Low risk |
Mohammad EE et al (2017) (21) | Unclear | Unclear | Unclear | Unclear | Low risk | Low risk | Low risk |
Shehata NA et al (2020) (22) | Low risk | Low risk | Low risk | Low risk | Low risk | Low risk | Low risk |
Two studies reported data on MCA RI, five on MCA PI, three on MCA S/D, three on UA RI, five on UA PI, four on UA S/D, and two on CPR. Table 3 shows the summary of pooled results for the effect of sildenafil citrate on ultrasound indices. The visual presentation of individual studies using forest plots for these outcomes is shown in Fig. 3, Fig. 4, Fig. 5, Fig. 6, Fig. 7, Fig. 8, Fig. 9, and Fig. 10. The pooled MD for improvement of MCA RI and MCA S/D was significantly higher in the sildenafil group than the comparison group (Pooled MD = 0.07, 95% CI:0.03 to 0.1, p-value < 0.001; and Pooled MD = 0.8, 95% CI: 0.18 to 0.45, p-value < 0.001, respectively). Although there was a higher pooled MD for MCA PI in the sildenafil citrate group than the comparison group, this difference was not statistically significant. Regarding the UA related outcomes, the women received sildenafil citrate had higher pooled MD for RI, and lower pooled MD for PI and S/D than the comparison group; however, these effects was not statistically significant (Pooled MD = 0.07, 95% CI: -0.05 to 0.19, p-value = 0.26, Pooled MD=-0.08, 95% CI: -0.23 to 0.07, p-value = 0.28, and Pooled MD=-0.02, 95% CI:-0.45 to 0.41, p-value = 0.93, respectively). No significant difference occurred in the mean value of CPR for women who received sildenafil citrate and those who did not (Pooled MD = 0.29, 95% CI: -0.22 to 0.81, p-value = 0.27).
Table 3
The summary of pooled results for estimating the effect size of treatment
Outcome | Model | N of Studies | Effect size and 95% confidence interval | Heterogeneity |
Mean Difference | Lower limit | Upper limit | P-value for overall effect | P-value for heterogeneity | I-squared |
MCA RI | Fixed | 2 | 0.07 | 0.03 | 0.10 | 0.00 | 0.69 | 0.00 |
MCA PI | Random | 5 | 0.55 | -0.16 | 1.26 | 0.13 | < 0.01 | 99.28 |
MCA S/D ratio | Fixed | 3 | 0.80 | 0.45 | 1.16 | 0.00 | 0.63 | 0.00 |
UA RI | Random | 3 | 0.07 | -0.05 | 0.19 | 0.26 | < 0.01 | 96.96 |
UA PI | Random | 5 | -0.08 | -0.23 | 0.07 | 0.28 | < 0.01 | 95.78 |
UA S/D ratio | Random | 4 | -0.02 | -0.45 | 0.41 | 0.93 | < 0.01 | 93.89 |
CPR | Random | 2 | 0.29 | -0.22 | 0.81 | 0.27 | < 0.001 | 91.23 |
Abbreviations. CPR, cerebroplacental ratio; MCA, middle cerebral artery; grRI, resistive index; S/D, systolic to the diastolic ratio; UA, umbilical artery. |