Results
In this study 54 patients were prospectively enrolled. 48 women were diagnosed with HELLP syndrome while six women fulfilled six or more Swansea criteria and therefore were diagnosed with AFLP. The clinical characteristics of the study population are depicted in Table 1.
AFLP patients showed impressively high sFlt-1 levels up to 147170pg/ml (median: 57570pg/ml [range: 31609-147170pg/ml]) which were significantly higher than those found in patients with HELLP syndrome (9713pg/ml [1348-30781]pg/ml; p<0.001; Figure 1).
PLGF serum levels were higher in patients with AFLP than in those with HELLP syndrome (median: 197 pg/ml [range 127-487pg/ml] versus 40 pg/ml [9-644pg/ml], p<0.01; Figure 2).
sFlt-1/PLGF-ratios were not significantly different between AFLP and HELLP syndrome patients (192 [157-1159] versus 232 [3-948]; NS); (Figure 3).
ROC analysis was performed to assess the test performance of sFlt-1 and PLGF in discriminating between AFLP and HELLP syndrome. A sFlt-1 cut-off value of 31.2 ng/ml allowed differentiating between these two diseases with a sensitivity and specificity of 100%. For PLGF on the other hand the area under the ROC curve was 0.89 (95%CI 0.79-0.98; p<0.005). A PLGF cut-off value of 110 g/ml showed a sensitivity of 100% with a specificity of 81 % in distinguishing between AFLP and HELLP syndrome.
To correct for a potential gestational age-dependent bias regarding the angiogenic parameters patients with AFLP and HELLP syndrome were matched (±1 week) on a 1:2 basis. Compared to gestational age-matched HELLP syndrome patients median [range] sFlt-1 serum levels were still higher in AFLP patients (HELLP: 9209 pg/ml [2544-30781] versus AFLP: 57570 pg/ml [range: 31609-147170pg/ml]; p<0.001). Similarly, median PLGF serum concentrations were higher in AFLP patients when compared to women affected by HELLP syndrome; (AFLP: 197pg/ml [127-487pg/ml; n=6] versus HELLP: 57pg/ml [17-232 pg/ml; n=12], p<0.01). Since serum levels of both sFlt-1 and PLGF were elevated in patients with AFLP when compared with those affected by HELLP syndrome sFlt-1/PLGF-ratio were not significantly different between AFLP and HELLP syndrome patients (AFLP: 192 [157-1159] versus HELLP: 202 [17-832], respectively, p=NS). There was no overlap in sFlt-1 serum concentrations between theses matched groups, a sFlt-1 cut-off value of 31100 pg/ml differentiated perfectly between the two pregnancy complications.
Three out of six patients (50%) in the AFLP group showed sFlt-1 levels over the threshold of the test detection limit of 85.0 ng/ml. Performing dilution series with human serum 1:10, sFlt-1 levels up to 147170 pg/ml were detected.
Further analysis showed that no FGR was found in the off-springs of our AFLP cohort, whereas 48% of all off-springs following pregnancies affected by HELLP syndrome were considered as FGR. (Table 1)The rate of NICU admission was comparable between neonates following pregnancies with AFLP and HELLP syndrome, however AFLP off-springs showed significantly lower umbilical cord artery pH values than their peers following HELLP syndrome: AFLP: 7.23 (7.11-7.28 versus HELLP: 7.33 (7.12-7.38; p<0.01). Apgar scores as well as maternal age or BMI did not differ significantly between the two groups although 50% of the neonates in the AFLP group had an Apgar score at 5 minutes of <7.0 No significant difference was found in aminotransferases concentrations between women affected by HELLP syndrome and AFLP.