Prognosis of combined pulmonary fibrosis and emphysema: comparison with idiopathic pulmonary fibrosis alone

Search strategy

We performed systematic searches of the medical literature for articles published in electronic databases including PubMed, Embase, and the Cochrane central register of controlled trials prior to 31 March 2018 according to a standardized protocol. Search terms were ‘emphysema’ and ‘idiopathic pulmonary fibrosis’. In addition, the reference lists of all the relevant studies and reviews were also checked by hand.

Study selection

Two reviewers (CJ and QF) independently evaluated the inclusion and exclusion criteria, and references for eligibility were determined by both reviewers. Disagreements were resolved by consensus. Eligibility criteria were as follows: (1) studies evaluating CPFE versus IPF; (2) articles that provided relevant survival data, Kaplan–Meier survival curves, or both. Studies presenting insufficient data were excluded as were duplicates, non-English studies, conference abstracts, editorials, reviews, case reports, or small case series (less than five patients).

Data extraction and quality assessment

Data were systematically extracted from the selected studies and entered onto a data extraction form designed before beginning the study. Trial characteristics including first author, publication year, country, the sample size, proportion of male patients, mean age of included patients, meaning of CPFE, duration of follow-up, and SRs at 1 year, 3 years, and 5 years were recorded to allow for exploration of potential reasons for any heterogeneity detected between trial results. If possible, the raw value for the survival rate was recorded. When these rates were unavailable, the survival rates were estimated from survival curves. No attempt was made to include unpublished data. The methodological qualities of the eligible studies were evaluated with the Newcastle Ottawa Quality Assessment Scale (NOS). ⁷ NOS scores of 0–3, 4–6, and 7–9 were considered to indicate low, moderate, and high quality, respectively.

Data synthesis and analysis

The data were analyzed using Review Manager (version 5.3) software by the Cochrane Collaboration (Oxford, England). As primary outcomes, the variances of 1-year, 3-year, and 5-year SRs of CPFE versus IPF alone were expressed as a combined relative risk (RR) with a 95% confidence interval (CI). When the p value of the χ² test was more than 0.05 or I² was less than 50%, the fixed-effect model weighted by the Mantel–Haenszel method was used. Otherwise, the random effect model was applied in the case of significant heterogeneity. A p < 0.05 was considered statistically significant.

Study identification

The process of identifying eligible studies is schematically illustrated in Figure 1. A total of 1114 citations from the initial search were found, of which 19 studies were retrieved for further assessment after title and abstract evaluation. Among these papers, four that compared CPFE and pulmonary fibrosis and two that lacked extractable survivable data were excluded. Ultimately, 13 publications met the inclusion criteria and were included in the meta-analysis.^8–20

OPEN IN VIEWER

Figure 1.

Flow diagram of the study selection process in the meta-analysis.

Detailed characteristics of the included trials are shown in Table 1. The 13 studies contained a total of 1710 participants (603 in the CPFE group and 1107 in the IPF alone group) published between 2009 and 2017. Two of the trials were conducted in Europe,^18,20 two in North America,^8,11 and the remaining nine trials in east Asia.^{9,10,12–17,19} The number of participants in each study ranged from 17 to 365 individuals. The proportion of male patients varied between 69% and 100% in the CPFE group, and between 57% and 74% in the IPF alone group. The mean age of individuals ranged from 64.0 to 75.0 years in the CPFE group, and from 60.0 to 73.7 years in the IPF alone group. The duration of follow-up on survival was between 1 and 15 years. The NOS was used to evaluate the quality of the selected studies. All studies were graded as high quality, with scores ranging from 7 to 9 (Table 1).

OPEN IN VIEWER

Table 1.

Characteristics of included trials.

Study	Location	Group	Sample size, n	Male	Age (years)	Duration of follow-up on survival (years)	NOS
Mejía et al. 8	Mexico	CPFE IPF alone	31 79	30 49	67 ± 7 63 ± 10	6.6	9
Akagi et al. 9	Japan	CPFE IPF alone	26 33	23 22	65.1 ± 8.5 66.5 ± 9.2	15	9
Kurashima et al. 10	Japan	CPFE IPF alone	129 233	NA NA	NA NA	10	8
Ryerson et al. 11	USA	CPFE IPF alone	29 336	20 239	69.9 ± 8.7 69.0 ± 8.6	8	9
Ye et al. 12	China	CPFE IPF alone	70 55	68 38	64 ± 9 66 ± 8	4	8
Sugino et al. 13	Japan	CPFE IPF alone	46 62	43 46	71.4 ± 6.7 73.7 ± 6.3	8	9
Kim et al. 14	Korea	CPFE IPF alone	26 42	23 24	67.6 ± 2.2 68.2 ± 1.7	11.5	9
Sato et al. 15	Japan	CPFE IPF alone	55 45	53 38	71.8 ± 7.3 69.9 ± 7.1	5	9
Zhang et al. 16	China	CPFE IPF alone	87 105	76 66	66 ± 8.5 60 ± 4.3	5	8
Sato et al. 17	Japan	CPFE IPF alone	12 5	NA NA	NA NA	3.5	7
Papaioannou et al. 18	Greece	CPFE IPF alone	29 62	26 43	75 72	1	8
Kohashi et al. 19	Japan	CPFE IPF alone	34 13	NA NA	NA NA	6.8	9
Portillo et al. 20	Spain	CPFE IPF alone	29 37	29 26	71 ± 7 72 ± 10	10.5	9

Data are mean ± standard error of the mean.

CPFE, combined pulmonary fibrosis and emphysema; IPF, idiopathic pulmonary fibrosis; n, number of pairwise comparisons; NOS, Newcastle Ottawa Quality Assessment Scale.

1-year SR comparison of CPFE versus IPF alone

Twelve trials reported the 1-year SR comparison of CPFE versus IPF alone.^{8–15,17–20} There were no statistically significant differences in the 1-year SR between the CPFE group and the IPF alone group (RR = 0.98, 95% CI: 0.94–1.03, p = 0.47). Heterogeneity was not evident, as assessed by the statistics (I² = 19%, p = 0.26) (Figure 2).

OPEN IN VIEWER

Figure 2.

Forest plot of trials of CPFE versus IPF alone on relative risk of 1-year SR.

3-year SR comparison of CPFE versus IPF alone

Eleven trials reported the 3-year SR comparison of CPFE versus IPF alone.^{8–15,17,19,20} Statistically significant differences in the 3-year SR between the CPFE group and the IPF alone group were not observed (RR = 0.83, 95% CI: 0.68–1.01, p = 0.06). Heterogeneity was substantial, as assessed by the statistics (I² = 77%, p < 0.00001) (Figure 3).

OPEN IN VIEWER

Figure 3.

Forest plot of trials of CPFE versus IPF alone on relative risk of 3-year SR.

5-year SR comparison of CPFE versus IPF alone

Ten trials reported the 5-year SR comparison of CPFE versus IPF alone.^{8–11,13–16,19,20} The combined results of the trials revealed that the CPFE group had no significant differences in the 5-year SR compared with the IPF alone group (RR = 0.80, 95% CI: 0.59–1.07, p = 0.14). Heterogeneity was substantial, as assessed by the statistics (I² = 79%, p < 0.00001) (Figure 4).

OPEN IN VIEWER

Figure 4.

Forest plot of trials of CPFE versus IPF alone on relative risk of 5-year SR.

Publication bias

The funnel plots for publication bias appeared to be symmetrical (Figure 5). These results indicated no evidence of publication bias for 1-year, 3-year, and 5-year SRs compared with the IPF alone group.

OPEN IN VIEWER

Figure 5.

Funnel plot to assess for evidence of publication bias. (a) Funnel plot for the studies on 1-year SR; (b) Funnel plot for the studies on 3-year SR; (c) Funnel plot for the studies on 5-year SR.