The present meta-analysis aimed to determine the outcomes and effect on right ventricular (RV) function of surgical pulmonary valve replacement (PVR) in patients after repair of tetralogy of Fallot. The reported outcomes of surgical PVR in children and adults after tetralogy of Fallot repair were from relatively small observational studies. The PubMed database was searched from its inception to April 2009. Observational studies reporting on the following outcomes measures after surgical PVR were reviewed: early and late all-cause mortalities, the redo-PVR rate, and changes in the indexed RV volumes, ejection fraction, and QRS duration after PVR. Of the 305 citations screened, 15 met the criteria and were analyzed. The pooled early mortality rate (n = 595) was 2.1% (95% confidence interval [CI] 1.1% to 4.0%). The late mortality rate was 0.5%/patient-year (95% CI 0.2% to 0.8%/patient-year), and the redo-PVR rate was 1.9%/patient-year (95% CI 1.3% to 2.5%/patient-year). Data on RV volumes and ejection fractions were available from 5 studies (n = 141). The pooled mean difference in the indexed RV end-diastolic and end-systolic volume was −63 ml/m 2 (95% CI −55 to −72) and −37 ml/m 2 (95% CI −30 to −45), respectively. No significant changes in the pooled mean difference of the RV ejection fraction (95% CI −1% to 3%) or QRS duration (95% CI −10 to 1 ms) were observed. In conclusion, surgical PVR in patients after tetralogy of Fallot repair has been associated with low early and late mortality and significant decreases in RV volumes but no changes in the RV ejection fraction or QRS duration.
The present meta-analysis was conducted to determine the outcomes after surgical pulmonary valve replacement (PVR) and its effect on right ventricular (RV) volumes and ejection fraction and QRS duration in pediatric and adult patient populations after operative “repair” of tetralogy of Fallot (TOF).
Methods
We performed the present systematic review and meta-analysis using the framework proposed by the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. Relevant publications in English were identified by searching PubMed (National Library of Medicine) from its inception to April 2009. The terms used for the search were “tetralogy of Fallot” and “pulmonary valve replacement” or “homograft” or “autograft” or “Hancock” or “Contegra” or “valved conduit.” Observational studies reporting the outcomes after PVR were included, and case reports and review articles were excluded. Studies on percutaneous PVR were also excluded from the present review because the data on intermediate- and long-term outcomes remain limited. One of us (EWYC) screened all the abstracts for eligibility, and a second reviewer (YFC) independently assessed the eligibility for inclusion and exclusion. Disagreements were resolved between the 2 reviewers. The full texts of all eligible articles were retrieved for detailed review. Additionally, potential relevant studies were identified by a manual search of the reference lists of all the eligible studies. If required, the corresponding authors of the publications were queried for additional information. In the case of multiple publications arising from the same patient cohort, the most recent report was selected.
The extracted data included the study design, study period, patient demographics, surgical technique, follow-up duration, and outcomes. The outcome measures assessed were early and late all-cause mortality, valve complications and deterioration requiring redo-PVR, changes in QRS duration, and changes in RV volumes and ejection fractions, as determined by cardiovascular magnetic resonance imaging (CMRI). If an outcome parameter was not reported in an eligible study, the study was excluded from the analysis of that particular parameter. The studies were divided into pediatric (aged ≤18 years) and adult (aged >18 years) according to the mean age at PVR.
Fixed-effects meta-analyses were performed using the Mantel-Haenszel method. Odds ratios and weighted mean differences with 95% confidence intervals (CIs) were calculated for the binary and continuous variables, respectively. The heterogeneity of the studies was assessed using Cochran’s Q test and I 2 statistic. Funnel plots were constructed to evaluate study publication bias. All tests were 2-sided, and p <0.05 was considered statistically significant. The data were analyzed using Review Manager, version 5.0.18 (Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Denmark, 2008).
Results
A total of 305 studies were identified and screened, of which 290 were excluded ( Figure 1 ). The characteristics of the 15 studies that met the criteria and were included in this review are summarized in Table 1 .
| Publication | Patients (n) | PVR Types | Mean Follow-Up (y) | Mean Age (y) | Outcome Parameters Available for Analysis | ||||
|---|---|---|---|---|---|---|---|---|---|
| Early Mortality | Late Mortality | Redo-PVR | CMRI Data | QRS Duration | |||||
| Yemets et al, 1997 | 85 | H,B,M | 5.8 | 20 | + | + | + | ||
| Discigil et al, 2001 | 42 | H,B | 7.8 | 22 | + | + | + | ||
| Therrien et al, 2001 | 70 | B | 4.7 | 28 | + | ||||
| De Ruijter et al, 2002 | 16 | H,B | 4.6 | NA | + | + | + | ||
| Warner et al, 2003 | 36 | H,B | 6.7 | 15 | + | + | + | ||
| Lim et al, 2004 | 58 | H,B,M | 2.5 | 14 | + | + | + | + | |
| Borowski et al, 2004 | 18 | H | 1.3 | 24 | + | + | |||
| Therrien et al, 2005 | 17 | B | 1.75 | 34 | + | + | + | + | |
| Buechel et al, 2005 | 20 | H,B | NA | 14 | + | + | + | ||
| Oosterhof et al, 2006 | 158 | H | Median 4.2 | NA | + | + | + | ||
| Oosterhof et al, 2007 | 71 | H | Median 1.6 | Median 29 | + | + | |||
| Ghez et al, 2007 | 19 | H | 1.6 | 24 | + | + | + | + | |
| Graham et al, 2008 | 93 | H,B,M | Median 3 | 26 | + | + | + | ||
| Knirsch et al, 2008 | 16 | H,B | NA | 12 | + | + | |||
| Meijboom et al, 2008 | 17 | NA | 6.4 | 28 | + | + | |||
Of these 15 studies, 13 reported early mortality data (n = 595), 11 reported late mortality data (n = 559, 2,415 patient-years), 9 reported on patients undergoing redo-PVR (n = 524, 2,283 patient-years), 5 reported on changes in CMRI-derived RV parameters (n = 142), and 4 reported on changes in QRS duration (n = 213). Because Oosterhof et al reported the mortality data and CMRI parameters separately in 2 publications, both of these studies were included. Knirsch et al provided data only for RV end-diastolic volume but not RV systolic volume or pulmonary regurgitant fraction.
Table 2 lists the pooled data on the early mortality, late mortality, and redo-PVR rates. The pooled early mortality rate was 2.1% (95% CI 1.1% to 4.0%). The late mortality rate was 0.5%/patient-year (95% CI 0.2% to 0.8%), and the redo-PVR rate was 1.9%/patient-year (95% CI 1.3% to 2.5%). When the pooled data from the adult and pediatric series were compared, the pooled early mortality rate was greater in the pediatric series, the late mortality rate was similar between the adult and pediatric series, and the redo-PVR rate was greater in the adult series.
| Publication | Early Mortality (%) | Late Mortality (%/patient-year) | Redo-PVR (%/patient-year) |
|---|---|---|---|
| Adult series | |||
| Yemets et al, 1997 | 1.2 (0.2–7.9) | 0.6 (0.0–1.3) | 1.6 (0.5–2.7) |
| Discigil et al, 2001 | 2.4 (0.3–15.1) | 1.8 (0.4–3.3) | 2.4 (0.7–4.1) |
| Borowski et al, 2004 | 5.6 (0.8–30.7) | 2.1 (0.0–7.9) | — |
| Therrien et al, 2005 | 2.8 (0.2–32.2) | 1.7 (0.0–6.2) | 1.7 (0.0–6.2) |
| Oosterhof et al, 2006 | 0.3 (0.0–4.8) | 0.3 (0.0–0.7) | 2.4 (1.2–3.6) |
| Ghez et al, 2007 | 2.5 (0.2–29.8) | 1.6 (0.0–6.1) | 3.3 (0.0–9.7) |
| Graham et al, 2008 | 0.5 (0.0–7.9) | 0.7 (0.0–1.7) | 3.6 (1.4–5.8) |
| Meijboom et al, 2008 | 2.8 (0.2–32.2) | 0.5 (0.0–1.7) | — |
| Pooled adult series | 1.8 (0.8–4.1) | 0.5 (0.2–0.8) | 2.2 (1.5–2.9) |
| Pediatric series | |||
| De Ruijter et al, 2002 | 6.3 (0.8–33.5) | 1.4 (0.0–4.0) | 1.4 (0.0–4.0) |
| Warner et al, 2003 | 1.4 (0.1–18.3) | 0.4 (0.0–1.2) | 0.8 (0.0–2.0) |
| Lim et al, 2004 | 1.7 (0.0–11.2) | 0.3 (0.0–1.3) | 4.1 (0.8–7.4) |
| Buechel et al, 2005 | 2.4 (0.2–28.7) | — | — |
| Knirsch et al, 2008 | 2.9 (0.2–33.6) | — | — |
| Pooled pediatric series | 2.7 (0.9–7.5) | 0.4 (0.0–1.0) | 1.2 (0.2–2.2) |
| Pooled total | 2.1 (1.1–4.0) | 0.5 (0.2–0.8) | 1.9 (1.3–2.5) |
| Heterogeneity test | |||
| Chi-square | 5.62 | 5.64 | 9.0 |
| p Value | NS | NS | NS |
| I 2 | 0% | 0% | 0% |
The effects of PVR on the CMRI-derived RV volumes, RV ejection fraction, pulmonary regurgitant fraction, and QRS duration are shown in Figure 2 . PVR was associated with a significant reduction in the indexed RV end-diastolic volume (mean difference −63 ml/m 2 , 95% CI −55 to −72), end-systolic volume (mean difference −37 ml/m 2 , 95% CI −30 to −45), and pulmonary regurgitant fraction (mean difference −38%, 95% CI −35% to −41%). In contrast, PVR did not result in significant changes in the RV ejection fraction (mean difference 1%, 95% CI −1% to 3%) and QRS duration (mean difference −4 ms, 95% CI −10 to 1).
