Skip to main content

Table 4 Summary of heart valve 16S ribosomal PCR results from patients with definite IE according to Duke’s criteria: literature review

From: Development and evaluation of a novel fast broad-range 16S ribosomal DNA PCR and sequencing assay for diagnosis of bacterial infective endocarditis: multi-year experience in a large Canadian healthcare zone and a literature review

Study (Reference) Location Median duration of pre-operative antibiotic treatment (days) Cases/Controlsa Cases with positive pre-operative blood culture(s) Nucleotide position of primers in E. coli 16S rRNA gene Agreement of HV culture and 16s PCR HV culture sensitivity/specificity Sensitivity Specificity PPV NPV
1 (Goldenberger, [8]) Switzerland NR b 18 14/18 (77.8 %) 8-806 2/18 (11.1 %) 11.1/100 % 93.3 % 66.7 % 93.3 % 66.7 %
2 (Gauduchon, [13]) France 31.5 (range, 8 to 150) 29/23 c 21/29 (72.4 %) 911-930 and 1390-1371 27/29 (93.1 %) c NR ND ND ND ND
3 (Lang, [19]) UK NR d 28/61 20/28 (71.4 %) 1522-1540 and 1170-1189 14/20 (70 %) NR ND ND ND ND
4 (Breitkopf, [23]) Germany NR 51/16 7/21 (33.3 %) 8-27 and 907-926 None 7.8/93.7 % 41.2 % 100 % 100 % 34.8 %
5 (Greub, [15]) France NR 127/118 57/127 (44.9 %) 536-1050 14/68 (20.6 %) 13/98 % 61 % 100 % 100 % 74 %%
6 (Rovery, [24]) France 19.5 (range, 1–150) 147 NR 536-1050 64/95 (67 %) e NR ND ND ND ND
7 (Kotilainen, [18]) Finland 19.6 (range 1–58d) 28/18 20/28 (71.4 %) 1054-1077 and 1950-1926 18/25 (72 %) 13.1/100 % 43.1 % 100 % 100 % 58.1 %
8 (Marin, [9]) Spain 10 (range, 1–25) 35/120 31/35 (88.6 %) 783-806 and 1389-1370 16/35 (45.7 %) 24.3/56.4 % 96 % 95.3 % 98.4 % 88.5 %
9 (Volstedlund, [25]) Denmark 19.3 (range, 0–90) 57/10 50/57 (87.7 %) 341-534 19/57 (33.3 %) 26/62 % 72 % 100 % 100 % 62 %
10 (Fournier, [30]) g France NR 549/191/19 All patients had negative blood culture Same as study (2) 536F and RP2 45.6 % 45.7 %/NR 69.2 % ND ND ND
11 (Miyazato, 2011) Japan   19 15/19 (79 %) 8UA and 1485B None: 79 % had a positive Gram stain None were positive 100 % 100 % 79 % 100 %
12 (Vondracek, [27]) Sweden NR 57/61 48/57 (84 %) 334-939 20/57 (35.1 %) 23/87 % 77 % 100 % 100 % 87 %
13 (Boussier, [28]) f France NR 31 23/31 (74.2 %) Same as (2) and 8-27 and 1510-1492 5/31 (16.1 %) 32.3/100 % 78 % 100 % 100 % 61.5 %
14 (Kemp, [17]) Denmark NR 56/36 36/56 (64.3 %) 8-534 7/42 (16.7 %) 16.7/100 % 88 % 61.5 % 88 % 57.1 %
15 (Sadaka, [29]) g Egypt NR 19 All blood cultures were negative 1522-1540 and 1170-1189 5/6 (83.3 %) 62.5 %/NR ND ND ND ND
16 (Harris, [16]) Ireland, UK NR 47 35/47 (74.5 %) 16S Fa, 16S Fb and 16SR (320 bp) 29/47 (61.7 %) NR 67 % 91 % 96 % 46 %
17 (Leli, [31])h Italy NR 20 NR Septifast (Roche) 3/19 (15.8 %) 15.8/100 % 95 % 100 % 100 % 83.3 %
18 (Marsch, [32]) i Germany NR 46 NR UMD™, Molzym 27/46 (56.7 %) 32.1/100 % 61 % ND ND ND
  1. aOnly data for definite IE cases is included. Some studies (3, 7, and 9) also enrolled a small number of cases with possible IE according to Duke’s criteria. Controls were patients undergoing cardiac surgery for non-infective reasons
  2. bNot reported (NR)
  3. cDefinite IE was diagnosed by histology in the 52 cases; 29 definite IE cases were included and 23 cases with no evidence of IE on histology of heart valve tissue. PCR results were compared to histology and not culture in this study
  4. dCases divided into an active group (n = 19) that was on antimicrobial therapy at the time of surgery, and the resolved group (N = 9) who had completed treatment (time period 3 months to 7 years after treatment at the time of cardiac surgery, and nine others with possible IE
  5. eBacterial DNA was amplified by PCR significantly more often (64/95, 67 %) in HV with histological evidence of IE than valves that had no histological evidence of IE (21/55, 38 %) (p = 0.001)
  6. fThis study used a two-step PCR procedure: first, a real-time method, then a conventional end-point PCR was applied to HV samples to improve the sensitivity of molecular detection from 38 to 58 %
  7. gBoth of these studies only enrolled patients with culture negative endocarditis (i.e., all patients had clinical evidence of IE but negative blood cultures)
  8. hPCR on heart valve tissue was performed using a commercial real-time PCR assay (SeptiFast, Roche Molecular Systems, Mannheim Germany)
  9. iPCR of heart valve tissue was performed using a commercial assay (UDM™, Molyzm, Bremen, Germany) that uses both 16S rRNA bacterial primers and 18S rRNA fungal primers