The mortality increase in cardiac surgery patients associated with shrunken pore syndrome correlates with the eGFR_{cystatin C}/eGFR_creatinine-ratio

Figures & data

Table 1. Preoperative characteristics of the study cohort (n= 4007) presented as either means (± standard deviation) for continuous variables or numbers (percent) for dichotomous variables.

Variable	Mean (±SD)/no. (%)
Age (years)	68.2 ± 10.3
Diabetes	994 (24.8%)
History of cerebrovascular injury	376 (9.4%)
COPD	451(11.3%)
Anaemia	339 (8.5%)
Previous vascular surgery	146 (3.6%)
Previous myocardial infarction	2060 (51.4%)
Female gender	935 (23.3%)
Previous vascular disease	405 (10.1%)
LVEF 30–50%	859 (21.4%)
LVEF <30%	255 (6.4%)
Euroscore I-additive	4.7 ± 3.0
IABP before surgery	47 (1.2%)
NYHA class	–
I	30.0%
II	34.0%
III	27.8%
IIII	8.2%
CCS class
I	21.6%
II	47.9%
III	34.9%
IIII	11.8%
eGFR (mL/min/1.73 m^2) according to:
CKD-EPIcreatinine	74.9 ± 20.0
CKD-EPIcystatin C	63.8 ± 21.7
CAPA	66.2 ± 20.9
LMrev	66.7 ± 17.5

CAPA: Caucasian Asian Pediatric Adult; CCS: Canadian Cardiovascular Society; CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration; COPD: chronic obstructive pulmonary disease; IABP: intra-aortic balloon pump; LMrev: Lund Malmö Revised; LVEF: left ventricular ejection fraction; NYHA: New York Heart Association.

Table 2. Postoperative characteristics of the study cohort (n= 4007) presented as either means (± standard deviation) for continuous variables or numbers (percent) for dichotomous variables.

Outcomes	Mean (±SD)/no. (%)
Use of CPB	4007 (100%)
Perfusion time (min)	80.4 (±37.6)
Aortic cross clamp time (min)	51.0 (±24.4)
IABP after surgery	59 (1.5%)
Postoperative RRT	48 (1.3%)
Myocardial infarction	24 (0.6%)
Pneumonia	49 (1.3%)
Sepsis	19 (0.5%)
Permanent stroke	12 (0.3%)
Atrial fibrillation	889 (23.9%)
Reoperation for bleeding	73 (2.1%)
Reoperation for DSWI	26 (0.6%)

CPB: cardiopulmonary bypass; DSWI: deep sternal wound infection; IABP: intra-aortic balloon pump; RRT: renal replacement therapy

Figure 1. One- and 3-year mortality in percentage at different ratios of CKD-EPI_{cystatin C}/CKD-EPI_creatinine. Error bars indicate standard error of the mean. Horizontal bars indicate statistical significance in Chi-square tests. CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration.

Table 3. Mean eGFR expressed in mL/min/1.73 m² ± SD and 1- and 3-year mortality (numbers and percentages) at different ratios of CKD-EPI_{cystatin C}/CKD-EPI_creatinine.

Ratio	n	Mean eGFR CKD-EPIcreatinine	Mean eGFR CKD-EPIcystatin C	1-year mortality (%)	3-year mortality (%)	Chi-square 1-year p value	Chi-square 3-year p value
<0.5	94	67.2 (± 18.8)	29.5 (± 8.8)	12 (12.8)	27 (28.7)	.287 .138 .001 .920 .352 .787 .303	.037 .073 .001 .934 .033 .231 .441
0.5–0.6	205	71.7 (± 21.1)	39.9 (± 12.1)	18 (8.8)	37 (18.0)
0.6–0.7	506	75.0 (± 19.0)	49.0 (± 12.6)	29 (5.7)	65 (12.8)
0.7–0.8	809	74.6 (± 20.1)	56.1 (± 15.3)	19 (2.3)	48 (5.9)
0.8–0.9	879	75.1 (± 21.1)	63.8 (± 18.0)	20 (2.3)	53 (6.0)
0.9–1.0	682	79.0 (± 18.9)	74.8 (± 18.3)	11 (1.6)	25 (3.7)
1.0–1.1	494	77.7 (± 19.0)	81.2 (± 19.8)	7 (1.4)	12 (2.4)
1.1–1.2	201	69.9 (± 15.3)	79.9 (± 20.6)	1 (0.5)	3 (1.5)

CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration; eGFR: estimated glomerular filtration rate

Figure 2. One- and 3-year mortality in percentage at different ratios of CAPA/LMrev. Error bars indicate standard error of the mean. Horizontal bars indicate statistical significance in Chi-square tests. CAPA: Caucasian Asian Pediatric Adult; LMrev: Lund Malmö Revised.

Table 4. Mean eGFR expressed in mL/min/1.73 m² ± SD and 1- and 3-year mortality (numbers and percentages) at different ratios of CAPA/LMrev.

Ratio	n	Mean eGFR CAPA	Mean eGFR LMrev	1 year mortality (%)	3 year mortality (%)	Chi-square 1 year p value	Chi-square 3 year p value
<0.6	94	27.4 (± 14.1)	52.4 (± 24.9)	11 (11.7)	25 (26.6)	.486 .432 .008 .004 .052 .477 .464 .587	.136 .154 .003 .004 .376 .503 .195 .346
0.6–0.7	156	42.4 (± 13.1)	64.6 (± 19.8)	14 (9.0)	29 (18.6)
0.7–0.8	372	50.9 (± 15.2)	67.1 (± 19.9)	26 (7.0)	51 (13.7)
0.8–0.9	633	59.0 (± 14.3)	69.1 (± 16.7)	21 (3.3)	50 (7.9)
0.9–1.0	856	65.2 (± 15.7)	68.5 (± 16.5)	10 (1.2)	37 (4.3)
1.0–1.1	780	70.7 (± 17.3)	67.3 (± 16.3)	19 (2.4)	41 (5.3)
1.1–1.2	540	76.8 (± 18.8)	67.0 (± 16.3)	10 (1.9)	24 (4.4)
1.2–1.3	331	80.4 (± 19.7)	64.8 (± 15.9)	4 (1.2)	9 (2.7)
1.3–1.4	150	81.9 (± 22.7)	60.9 (± 16.8)	1 (0.7)	2 (1.3)

CAPA: Caucasian Asian Pediatric Adult; LMrev: Lund Malmö Revised; eGFR: estimated glomerular filtration rate

Figure 3. ROC curve plotting ratios of the equation pair CKD-EPI_{cystatin C}/CKD-EPI_creatinine (blue line) and the equation pair CAPA/LMrev (red line) for 1-year mortality. CAPA: Caucasian Asian Pediatric Adult; CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration; LMrev: Lund Malmö Revised; ROC: receiver operating characteristic.

Table 5. Four-field-matrices showing the overlaps for the two equation pairs used for diagnosing SPS.

		CAPA/LMrev
		No SPS (>0.85)	SPS (<0.85)
CKD-EPI	No SPS (>0.75)	2786	40
	SPS (<0.75)	313	868
		No SPS (>0.69)	SPS (<0.69)
CKD-EPI	No SPS (>0.57)	3757	31
	SPS (<0.57)	34	185

CKD-EPI refers to the CKD-EPI_{cystatin C}/CKD-EPI_creatinine-ratio with cut-offs set at 0.75 (upper matrix) and 0.57 (lower matrix). CAPA/LMrev refers to the CAPA/LMrev-ratio, with cut-offs set at 0.85 (upper matrix) and 0.69 in (lower matrix). The cut-offs used in the upper matrix represent Youden indexes calculated from ROC-curves, whereas the cut-offs used in the lower matrix were chosen to produce 95% specificity.

CAPA: Caucasian Asian Pediatric Adult; CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration; LMrev: Lund Malmö Revised; SPS: shrunken pore syndrome

Table 6. Four-field-matrices showing 1- and 3-year mortalities in percentages separated by a slash.

		CAPA/LMrev
		No SPS (>0.85)	SPS (<0.85)
CKD-EPI	No SPS (>0.75)	1.5%/4.1%	0%/0%
	SPS (<0.75)	3.2%/7.3%	7.4%/15.1%
		No SPS (>0.69)	SPS (<0.69)
CKD-EPI	No SPS (>0.57)	2.4%/5.8%	6.5%/9.7%
	SPS (<0.57)	2.9%/17.6%	12.4%/24.9%

CKD-EPI refers to the CKD-EPI_{cystatin C}/CKD-EPI_creatinine-ratio with cut-offs set at 0.75 (upper matrix) and 0.57 (lower matrix). CAPA/LMrev refers to the CAPA/LMrev-ratio, with cut-offs set at 0.85 (upper matrix) and 0.69 in (lower matrix). The cut-offs used in the upper matrix represent Youden indexes calculated from ROC-curves, whereas the cut-offs used in the lower matrix were chosen to produce 95% specificity.

CAPA: Caucasian Asian Pediatric Adult; CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration; LMrev: Lund Malmö Revised; SPS: shrunken pore syndrome.

Figure 4. ROC curve plotting ratios of the equation pair CKD-EPI_{cystatin C}/CKD-EPI_creatinine (blue line) and the equation pair CAPA/LMrev (red line) with 3-year mortality. CAPA: Caucasian Asian Pediatric Adult; CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration; LMrev: Lund Malmö Revised; ROC: receiver operating characteristic.