J Nephrol DOI 10.1007/s40620-014-0138-0
ORIGINAL ARTICLE
Survival after arterial embolization therapy in patients with polycystic kidney and liver disease Junichi Hoshino • Tatsuya Suwabe • Noriko Hayami • Keiichi Sumida Koki Mise • Masahiro Kawada • Aya Imafuku • Rikako Hiramatsu • Masayuki Yamanouchi • Eiko Hasegawa • Naoki Sawa • Ryoji Takei • Kenmei Takaichi • Yoshifumi Ubara
•
Received: 22 May 2014 / Accepted: 1 September 2014 Ó Italian Society of Nephrology 2014
Abstract Background Transcatheter arterial embolization (TAE) has become a therapeutic option for symptomatic polycystic kidney disease (PKD) and polycystic liver disease (PLD). However, factors affecting survival with renal TAE remain unknown. Methods All symptomatic patients with severe PKD and/ or PLD who received renal and/or hepatic TAE at our center from October 1996 through March 2013 (n = 1,028) were followed until death. Their survival was compared with that of the general PKD population on dialysis in Japan. Factors affecting survival were analyzed using the Cox hazard model. Results After renal TAE, 5- and 10-year survival was, respectively, 0.78 (95 % confidence interval, 0.74–0.82) and 0.56 (0.49–0.63); with hepatic TAE, 0.69 (0.58–0.77) and 0.41 (0.22–0.60); and with dual TAE (renal and hepatic), 0.82 (0.72–0.88) and 0.45 (0.31–0.59). Survival after dialysis initiation was better among patients with renal TAE than among general PKD patients. Factors Electronic supplementary material The online version of this article (doi:10.1007/s40620-014-0138-0) contains supplementary material, which is available to authorized users. J. Hoshino (&) T. Suwabe N. Hayami K. Sumida K. Mise K. Takaichi Y. Ubara Nephrology Center, Toranomon Hospital, 1-3-1, Kajigaya, Takatsu-ku, Kawasaki, Kanagawa 213-8587, Japan e-mail: [email protected] J. Hoshino T. Suwabe N. Hayami K. Sumida M. Kawada A. Imafuku R. Hiramatsu M. Yamanouchi E. Hasegawa N. Sawa K. Takaichi Y. Ubara Nephrology Center, Toranomon Hospital, Tokyo, Japan R. Takei Department of Radiology, Touzan Hospital, Tokyo, Japan
affecting survival after renal TAE were age [hazard ratio (HR) 3.02 (1.44–6.33) for every 10 years] and albumin [HR 0.70 (0.55–0.89) per 0.1 g/dl]. Kidney volume was not associated with patient death after TAE. The main causes of death among patients after renal TAE were similar to those of the general PKD population on dialysis whereas, after hepatic TAE, the main cause was cyst infection with liver failure (12.5 % with PLD and 5.9 % with PKD, p \ 0.01). Conclusion Survival after renal TAE with severe PKD was better than for the general PKD population on dialysis, suggesting that renal TAE could overcome the disadvantage due to huge organ size. Keywords Autosomal dominant polycystic kidney disease (ADPKD) Hepatic transcatheter arterial embolization (hepatic TAE) Renal transcatheter arterial embolization (renal TAE) Polycystic liver disease (PLD) Survival Cause of death
Introduction Autosomal dominant polycystic kidney disease (ADPKD) is the most common hereditary kidney disease [1, 2] and polycystic liver disease (PLD) its most common extrarenal manifestation [3]. Several studies reported that over 80 % of ADPKD patients had hepatic cysts [4, 5], risk factors for developing them being severity of renal cysts and poor renal function [5, 6]. Huge kidneys and livers may lead to malnutrition and can be lethal, with kidney and cyst volumes reportedly increasing exponentially, annual growth rate calculated at 5.27–5.36 % [7, 8]. Conservative treatment approaches are only partially effective and do not change the disease’s course [9, 10]. Accordingly, in
123
J Nephrol
October 1996 we initiated transcatheter arterial embolization (TAE) for treating symptomatic severe ADPKD/PLD patients (i.e., PKD and/or PLD with huge organ size), and noted mass volume reduction and nutritional improvement in both kidney and liver patients [11–15]. After renal TAE, over 50 % reduction in kidney volume and significant improvement in anemia and nutritional status were observed. With hepatic TAE, though—while mild reduction of cystic volume reduction and improvement of anemia were observed—the results were not as dramatic because the hepatic TAE embolized only cyst-occupied segments and hepatic parenchymal improvement was also seen with hepatic cyst contraction [15]. Hepatic TAE has become an accepted treatment for highly symptomatic patients who have large areas of cystic liver without recognizable hepatic parenchyma, and are not good candidates for combined liver resection, cyst fenestration, or liver transplantation [3, 13, 14, 16]. In the present longitudinal study, we aimed to clarify the rate of survival after renal and hepatic TAE, compare causes of death, and analyze the effectiveness of these treatments by comparing the survival with that reflected in the national data for Japanese dialysis patients (the general PKD/PLD population, some of whom may have received TAE unknown to us).
Subjects and methods Patients and study design From October 1996 through March 2013, 1,028 ADPKD patients received TAE treatment in our facilities. Diagnosis of polycystic kidney and liver was established by computed tomography (CT), magnetic resonance imaging, and ultrasonography with Ravine’s revised unified diagnostic criteria [17]. Inclusion criteria for renal TAE were: being on maintenance dialysis with daily urinary volume less than 500 ml/day; not being a good candidate for percutaneous cyst aspiration with/without sclerosis, or for surgery (or having declined other interventions); and having compression symptoms related to enlarged polycystic kidney or active renal cyst bleeding. Patients with uncontrollable renal cyst infection were excluded. Inclusion criteria for hepatic TAE—described previously [15]—were: having symptomatic PLD due to mass effects from hepatic segments almost completely occupied by cysts (over 90 % per segment), but with at least one segment almost intact, maintaining functioning liver parenchyma; and not being a candidate for surgery (or having declined other interventions). Exclusion criteria for hepatic TAE were: severe liver dysfunction (total bilirubin [2.0 mg/dl), massive ascites, diffusely
123
distributed intrahepatic cysts, or severe cyst infection associated with malnutrition. All TAE patients had expressed a strong desire for TAE therapy and gave consent after being fully informed about TAE’s risks. Of the 1,028 ADPKD patients, 68 % (n = 695, 45 % female) who had a markedly enlarged polycystic kidney received renal TAE, 21 % (n = 221, 77 % female) with marked PLD without a huge polycystic kidney received hepatic TAE, and 11 % (n = 112, 73 % female) having both an enlarged kidney and liver received dual TAE. In all— because of dual treatments—807 patients received renal TAE, and 333 received hepatic TAE. Looking at initial treatment (even if it was the only TAE received), 762 patients received renal TAE and 266 received hepatic TAE. With dual TAE, 67 of 112 (59.8 %) received renal TAE first. Interestingly, over 80 % [81 % in chronic kidney disease (CKD) G4-5, 85 % in CKD G1-3] of PLD patients with preserved renal function were female, suggesting that gender strongly affects PLD development. Patients’ demographic data were obtained at the time of TAE, cause of death from medical records. All laboratory data—including serum total protein (TP), serum albumin (Alb), serum creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), cholinesterase (ChE), c-glutamyl transpeptidase (GGT), C-reactive protein (CRP), hemoglobin (Hb), abdominal circumference, cardiac ejection fraction, and kidney/liver volume—were collected before treatment and 3, 6, and 12 months afterwards. Measurement of organ volume was described previously [13–15]. Organ volumes were measured from the set of contiguous images by summing the products of the organ areas traced in each CT image and the slice thickness. All laboratory values were measured by the automated, standardized methods used in our hospital within 24 h after drawing blood samples. The primary outcome of this study was death, with all patients followed until death or end of follow-up, using medical records, telephoning and/or writing to them. Methods of renal and hepatic TAE Methods of administering renal and hepatic TAE have been detailed previously [11–15]. To summarize, we performed renal and hepatic angiography using Seldinger’s technique, with a five-French shepherd hook catheter (Cathex, Shibuya, Tokyo, Japan) for renal angiography, and, for hepatic angiography, two catheters together (2- and 4-French) to navigate the hairpin-like meander of hepatic arterial branches. After the targeted arterial branches had been evaluated and selected, the arteries were carefully embolized from the periphery using platinum microcoils with a 0.46 mm diameter and 20–140 mm length (C-stopper coil, Piolax Medical Devices Inc., Yokohama, Japan; and
J Nephrol
TornadoÒ Embolization Coils, Cook Medical Inc., Bloomington, IN, USA). Usually, patients had fever and abdominal pain for a week, controllable with pentazocine, hydroxyzine hydrochloride, and nonsteroidal anti-inflammatory drugs. In renal TAE, an epidural catheter was inserted into the tenth or eleventh thoracic vertebra to control the severer pain of kidney embolization. Complications attending the procedure were pseudoaneurysm of the femoral artery (0.6 %) and the microcoil entering the aorta (0.8 %). Serious complications after renal TAE were emphysematous cyst infection (0.2 %) and disseminated intravascular coagulation (0.1 %). No serious treatment-related adverse effects were observed after hepatic TAE [15].
rank test. To eliminate the immortal person-time effect, we excluded patients who died within 2–4 years. The Cox proportional hazard model was used to obtain hazard ratios (HRs) and a 95 % confidence interval (CI) for death after TAE adjusted for sex, age, dialysis vintage, Alb, ALP, CRP, and kidney volume. We categorized patients into quartiles by original organ volume after treatment, dual TAE recipients divided by initial organ volume. All analyses used StataÒ SE version 12.0 (StataCorp, College Station, TX, USA).
Results Patients’ characteristics
Statistical analysis Data were summarized using proportions and means [±standard deviation (SD)] as appropriate. Categorical variables were analyzed with the Chi-squared or Fisher’s exact test, continuous variables compared using Student’s t test, Mann–Whitney U test, or analysis of variance (ANOVA). Cumulative survival was estimated with Kaplan–Meier survival curves, and compared using the logTable 1 Characteristics of ADPKD patients who received TAE
Table 1 and Fig. 1 show the breakdown and characteristics of ADPKD patients who received TAE. With hepatic TAE, mean TP, Alb, AST, ALT, ALP, GGT, and Hb were higher than in the other two groups (p \ 0.01) and mean age was lower (53.4 ± 9.2 years for hepatic TAE vs. 57.9 ± 9.0 for renal and 56.3 ± 6.8 for dual TAE, p \ 0.01). Median organ volumes were similar in all groups: median kidney volumes with renal and dual TAE were 4,667 and
Total (n = 1,028)
Renal TAE (n = 695)
Dual TAE (n = 112)
Hepatic TAE (n = 221)
p value
Sex (female %)
55.1 %
45.2 %
73.2 %
76.8 %
\0.01
Age (years)
56.7 ± 9.0
57.9 ± 9.0
56.3 ± 6.8
53.4 ± 9.2
\0.01 \0.01
Under dialysis (%)
85.3 %
100.0 %
96.4 %
33.5 %
Dialysis vintage (years)
5.8 ± 5.2
6.2 ± 5.3
3.9 ± 4.1
5.7 ± 5.1
0.02
Total protein (g/dl)
6.8 ± 0.6
6.7 ± 0.6
6.8 ± 1.0
7.0 ± 0.6
\0.01 \0.01
Albumin (g/dl)
3.1 ± 0.4
3.1 ± 0.4
3.1 ± 0.4
3.3 ± 0.4
AST (IU/l)
18 ± 16
16 ± 17
17 ± 7
23 ± 16
\0.01
ALT (IU/l)
13 ± 15
11 ± 13
12 ± 6
18 ± 19
\0.01 \0.01
ALP (IU/l)
287 ± 227
261 ± 220
275 ± 128
361 ± 250
ChE (IU/l)
174 ± 98
179 ± 95
167 ± 97
164 ± 105
0.16
GGT (IU/l) CRP (mg/dl)
81 ± 108 1.0 ± 2.8
57 ± 78 1.1 ± 3.2
86 ± 56 1.0 ± 2.9
142 ± 155 0.7 ± 1.1
\0.01 0.19
Hemoblobin (g/dl)
10.4 ± 1.6
10.2 ± 1.5
9.9 ± 1.2
11.0 ± 1.8
\0.01
6,433 (4,6389,463)
5,656 (4,3447,723)
0.3
Target organ volume Liver volume (cm3) Kidney volume (cm3) ADPKD autosomal dominant polycystic kidney disease, TAE transcatheter arterial embolization, AST aspartate aminotransferase, ALT alanine aminotransferase, ALP alkaline phosphatase, ChE cholinesterase, GGT c-glutamyl transpeptidase, CRP C-reactive protein
4,667 (3,2896,471)
5,468 (3,8806,256)
0.45
Abdominal circumference (cm)
92.1 ± 12.3
93.9 ± 10.4
89.8 ± 6.4
91.5 ± 13.4
0.4
Cardiac ejection fraction (%)
68 ± 10
67 ± 11
70 ± 9
71 ± 8
0.03
Mean TAE duration (year)
1.9 ± 2.1 2.0 ± 1.9 (liver first) (n = 45) 1.9 ± 2.2 (kidney first) (n = 67)
123
J Nephrol Fig. 1 Breakdown of ADPKD patients treated with TAE ADPKD autosomal dominant polycystic kidney disease; TAE transcatheter arterial embolization. Chronic kidney disease (CKD) G1-3: estimated glomerular filtration rate (eGFR) C30 ml/min/1.73 m2; G4, 5: eGFR \30 ml/min/ 1.73 m2
5,468 cm3, respectively (p = 0.45), median liver volumes with hepatic and dual TAE were 5,656 and 6,433 cm3, respectively (p = 0.30). Mean abdominal circumferences were also similar among the three groups (p = 0.40). The mean interval between sequential treatments with dual TAE was 1.9 ± 2.1 years (2.0 ± 1.9 with hepatic TAE first, and 1.9 ± 2.2 years with renal TAE first). Survival after TAE Comparing post-TAE survival curves, overall survival among the three groups was similar (p = 0.98), though slightly better with only renal TAE (Fig. 2a). With renal TAE, 3-, 5-, and 10-year survival was, respectively: 0.87 (95 % CI, 0.84–0.90), 0.78 (0.74–0.82), and 0.56 (0.49–0.63); with hepatic TAE, 0.86 (0.80–0.91), 0.69 (0.58–0.77), and 0.41 (0.22–0.60); and with dual TAE, 0.94 (0.88–0.97), 0.82 (0.72–0.88), and 0.45 (0.31–0.59). Next, we compared survival after renal TAE among patients with and without sequential hepatic TAE to determine the effect of hepatic TAE. For dual TAE patients who had renal treatment first, 3-, 5-, and 10-year survival was, respectively, 0.95 (0.85–0.98), 0.81 (0.68–0.90), and 0.42 (0.24–0.59) (Fig. 2b). That curve did not significantly differ from the curve for renal TAE (p = 0.29, log-rank). However, because mean duration between renal and hepatic TAE with dual treatment was 1.9 ± 2.2 years, one must exclude those surviving B2 years in order to minimize the effect of the duration of sequential TAE treatment
123
(i.e., immortal person-time). Considering that exclusion, survival with renal TAE was significantly better than with dual, kidney-first TAE (p = 0.03, log-rank), suggesting that the prognosis of patients with PLD was worse—even after hepatic TAE—than for patients who were nonsymptomatic or had no PLD (Fig. 2c). Then we compared survival after hepatic TAE among patients with and without sequential renal TAE to determine the effect of sequential renal TAE on those with huge kidneys. After dual TAE with hepatic TAE first, 3-, 5-, and 10-year survival was, respectively, 0.93 (0.80–0.98), 0.82 (0.64–0.91), and 0.51 (0.25–0.71) (Fig. 2d). Although overall survival after hepatic TAE without renal TAE seemed inferior to that with renal TAE (p = 0.23, logrank), again, it was similar between the two groups for those surviving [2 years (p = 0.83, log-rank, Fig. 2e). Considering the immortal person-time effect, the prognosis with severe PKD after renal TAE was very close to that for non-symptomatic PKD patients. We analyzed factors affecting survival after renal TAE with Cox models (Table 2). In the adjusted model, HRs Fig. 2 Overall survival after TAE a overall survival among patients c who received renal transcatheter arterial embolization (TAE), hepatic TAE, or dual TAE (sequential renal and hepatic TAE). b Overall survival after renal TAE with or without hepatic TAE. The mean duration of renal TAE and hepatic TAE with dual TAE was 1.9 ± 2.2 years. c Survival after renal TAE for those surviving more than 2 years. d Overall survival after hepatic TAE with or without renal TAE. The mean duration of hepatic TAE and renal TAE with dual TAE was 2.0 ± 1.9 years
J Nephrol
123
J Nephrol Table 2 Hazard ratios of all causes of mortality after renal TAE
Model 1 (unadjusted) HR
95 % CI-L
95 % CI-H
Model 2 (adjusted) p value
HR
95 % CI-L
95 % CI-H
p value
0.12
Kidney volume (cm3)
Model was adjusted for TAE type, sex, age, Alb, ALP, CRP, dialysis vintage, and target organ volume TAE transcatheter arterial embolization, Alb serum albumin, ALP alkaline phosphatase, CRP C-reactive protein, HR hazard ratio, CI confidence interval
Quartile 1 (mild)
1.00
Quartile 2
0.79
0.35
1.79
0.57
0.24
0.04
1.44
Quartile 3
0.36
0.13
1.00
0.50
0.63
0.12
3.36
0.60
Quartile 4 (severe)
1.04
0.50
2.15
0.93
1.26
0.27
5.87
0.77
Male sex
1.41
1.05
1.91
0.02
0.50
0.15
1.67
0.26
Dialysis vintage B5 years
1.00
1.00
5–10 years
0.99
0.62
1.60
0.98
0.64
0.14
2.87
0.56
[10 years
2.19
1.38
3.48
0.001
0.46
0.10
2.03
0.31
Age (every 10 years older)
1.89
1.56
2.29
\0.001
3.02
1.44
6.33
0.003
Alb (per 0.1 g/dl)
0.87
0.83
0.91
\0.001
0.70
0.55
0.89
0.004 0.36
ALP (per 100 IU/l)
1.20
1.15
1.25
\0.001
1.09
0.91
1.30
CRP (mg/dl)
1.04
1.01
1.07
0.01
1.03
0.85
1.24
0.78
Dual treatment with hepatic TAE
1.13
0.79
1.61
0.51
0.15
0.01
2.72
0.20
among kidney-volume groups were not significant, though the HR in patients with the largest organ volume—compared to that of those with the smallest volume—was 1.26 (0.27–5.87). Age [HR 3.02 (1.44–6.33) for every 10 years of patient age] and Alb [HR 0.70 (0.55–0.89) per 0.1 g/dl] were significant survival factors. Survival after initiation of dialysis with/without TAE The next question was whether TAE improved survival for its recipients compared to survival of the general PKD population on dialysis in Japan as reported annually by the Japanese Society of Dialysis Therapy (JSDT) [18]. According to that report, average 3-, 5-, 10- and 15-year survival for the Japanese PKD population on hemodialysis (henceforth, ‘‘JSDT population’’) was, respectively, 0.86, 0.77, 0.58, and 0.41. In comparison, survival after initiation of dialysis was significantly better with renal TAE than for the other two TAE/dialysis groups and the JSDT population (p \ 0.01, log-rank), but survival was similar with hepatic-only and dual TAE, and for the JSDT population. For instance, with renal TAE, 5-year survival was 0.97 (0.94–0.98) vs. the JSDT population’s 0.77, whereas with only hepatic TAE it was 0.78 (0.66–0.86) and with dual TAE, 0.85 (0.76–0.91). Figure 3 charts each time point. Since mean dialysis vintage at the time of TAE was 6.2 ± 5.3 years with renal TAE, 5.7 ± 5.1 years with hepatic TAE, and 3.9 ± 4.1 years with dual TAE, we next calculated patients’ survival after initiation of dialysis with 6-year censoring for renal and hepatic TAE, but 4-year
123
1.00
censoring for dual TAE group to minimize the immortal person-time effect. Again, survival with renal TAE was significantly better than for the JSDT population (Fig. 3a), though survival with hepatic TAE was a little worse at 10 years and only a bit better at 15 years. These findings suggested that renal TAE could overcome severe PKD’s disadvantage of malnutrition, whereas hepatic TAE’s effect for severe PLD was equivocal (Fig. 3b). In addition, with 4-year censoring, survival for dual TAE after initiation of dialysis was very similar to that of the general population (Fig. 3c). Causes of death Table 3 shows causes of death for TAE patients. Interestingly, the causes of death with renal and hepatic TAE were quite different. With renal TAE, the main causes were cardiovascular (3.9 % of all the patients in the treatment group) and cerebrovascular (2.4 %), much as with the JSDT population [18], while the death rate from liver failure resulting from liver cyst infection was significantly higher with hepatic TAE (5.9 %) and dual TAE (12.5 %) than with renal TAE (1.7 %) (p \ 0.01); comparison of dual TAE’s and hepatic TAE’s death rates from liver failure (5.9 % vs. 12.5 %) was p = 0.054 by Fisher’s exact test. The rate of other infectious deaths was slightly higher for dual TAE (2.7 %) than for renal (1.0 %) and hepatic (9 %) (p = 0.15, Fisher’s exact test). Note that the proportion of death with dual TAE was significantly higher than in the other two groups (34.8 vs. 20.3 % renal and
J Nephrol Fig. 3 Overall survival after initiation of dialysis. a overall survival among patients on dialysis who received renal transcatheter arterial embolization (TAE), hepatic TAE, or dual TAE (sequential renal and hepatic TAE). Survival data for the Japanese PKD population on dialysis is from the annual report (as of 31 December 2011) published by the Japanese Society for Dialysis Therapy. *It was p \ 0.01 between renal TAE and other groups including the general PKD/PLD population. It was p = 0.07 between the dual TAE and hepatic TAE groups. b and c Survival after initiation of dialysis for those surviving more than 4 and 6 years. The mean dialysis vintage at the time of TAE was 6.2 ± 5.3 years with renal TAE, 5.7 ± 5.1 years with hepatic TAE, and 3.9 ± 4.1 years with dual TAE
19.5 % hepatic, p \ 0.01), suggesting poorer prognosis in patients with severe PKD and PLD.
Discussion We studied the survival of Japanese patients with severe PKD and/or PLD after TAE. In Japan, ADPKD is the fourth leading cause of primary disease for patients on maintenance dialysis (3.4 %) [19]. Yet treatment options for severe symptomatic PKD and PLD remain very limited. Previously, we reported organ volume reduction and nutritional improvement with renal and hepatic TAE [11– 15]. As a result, more than 1,000 patients have come to our
hospital wanting TAE. Their median kidney volume was 4,667 cm3, over two SD larger than the mean kidney volume reported in Japan [20]—1,839 ± 1,329 cm3—suggesting the poorer general condition of our cohort. Based on our analyses, five things seem certain. First, 5and 10-year survival of patients after renal TAE was, respectively, 78 and 56 %, and after hepatic TAE 69 and 41 %. Second, outcome for patients after renal TAE was better than for patients without renal TAE—even with the immortal person-time effect—suggesting that renal TAE can overcome the disadvantage of enlarged kidneys. Third, the effect of hepatic TAE may be limited: in our analyses, survival after hepatic TAE was inferior to that of patients without symptomatic PLD; so hepatic TAE may not be
123
J Nephrol Table 3 Causes of death after TAE
Renal
TAE
Hepatic
TAE
221
Dual
TAE
112
Total
Number of patients
695
Total deaths
141
20.3 %
43
19.5 %
39
34.8 %a,b
1,028 223
Cyst infection/liver failure
12
1.7 %
13
5.9 %a
14
12.5 %a
39
3.8 %
Infection (pneumonia etc.)
7
1.0 %
2
0.9 %
3
2.7 %
12
1.2 %
21.7 % 0.0 %
The number of patients who died of each cause is shown, as well as the percentage of total cohort deaths each number represents
Malignancy
12
1.7 %
2
0.9 %
1
0.9 %
15
1.5 %
Cardiovascular
27
3.9 %
4
1.8 %
3
2.7 %
34
3.3 %
Cerebrovascular
17
2.4 %
4
1.8 %
3
2.7 %
24
2.3 %
Respiratory failure
1
0.1 %
0
0.0 %
0
0.0 %
1
0.1 %
a
Sudden death
9
1.3 %
0
0.0 %
1
0.9 %
10
1.0 %
p \ 0.01, comparing to renal TAE b
p \ 0.01, comparing to hepatic TAE
Other (accident etc.)
17
2.4 %
5
2.3 %
6
5.4 %
28
2.7 %
Unknown
39
5.6 %
13
5.9 %
8
7.1 %
60
5.8 %
enough to overcome the survival disadvantage of severe PLD. Fourth, the trend in causes of death differed between patients with renal and hepatic TAE. The main cause of death with severe PLD was liver cystic infection leading to liver failure, whereas for patients with huge kidneys the main causes were cardiovascular and cerebrovascular disease, the same as for the JSDT population. Since infections are reportedly the second most common cause of ADPKD patient death [21]—and most cyst infection occurs in livers [22]—the difference in causes of death with hepatic TAE may be reasonable. Nonetheless, the low percentage of infectious death with renal TAE seems notable—perhaps partly attributable to the marked nutritional improvement after renal TAE. Finally, although gender distribution of ADPKD patients with enlarged kidneys was almost equal, over 80 % of ADPKD patients with an enlarged liver that had preserved renal function were female, suggesting a different mechanism of cystic growth for renal and hepatic cysts: estrogen reportedly affects hepatic cyst growth, but not renal cystic growth [23]. Perhaps female hormonal change decreases the gender difference in prevalence of severe PLD patients on dialysis. There are some limitations to this study. First, referral bias may limit the external validity of our results. However, according to a hospital-based study of ADPKD prevalence in Japan [1], our cohort appears representative of Japanese ADPKD patients with severe PKD and PLD. Moreover, the mean age of our population and of those in that report was similar (56.7 ± 9.0 and 53.3 ± 15.9 years), so the problem of external validity may be insignificant. Second, this study did not have control patients—and could not—because all patients had severe conditions and were contraindicated for other therapies or were referred to our hospital because of a strong desire to receive TAE. Comparing survival with renal or hepatic TAE versus that with dual TAE (or of the JSDT population) could minimize this limitation. Third, our patients were younger than the JSDT population (their
123
mean age, 56.7 ± 9.0 vs. 64.4 ± 11.2 years). However, survival curve trends were very similar to our initial findings if we compared only our age-matched patients (64.3 ± 4.9 years) with the JSDT population (supplementary Fig. 1). Fourth, improvement of treatment technologies may confound results in long-term follow-up studies. However, survival after renal TAE in our study was similar in different periods: 5-year survival was 77.4 % in 1996–2000, 80.1 % in 2001–2005, and 73.7 % 2006–2008, implying that treatment year may not have had a large impact on our analyses. Finally, more experiences with selective renal/hepatic artery embolization by other groups and/or randomized prospective studies are needed to confirm the safety and efficacy of this procedure. Although TAE is increasingly recognized in Japan and Korea as a non-invasive option for highly symptomatic PKD/PLD patients who are not good candidates for surgical treatment [11–14, 24], TAE in lieu of surgery is not yet established as part of the medical culture in Western countries. The reason for this difference could be the higher rate of kidney/liver transplantation in the West, with cystic kidneys removed at the time of transplantation. In our study, 5-year mortality after dual TAE was 15 %, comparable to the expected 20 % 5-year mortality for combined liver/kidney transplantation in PLD patients [25]. In addition, 5-year mortality after hepatic TAE was 22 % in our study. It was reported that for PLD patients in better general condition than ours, overall mortality of traditional invasive approaches was 2 % with fenestration, 3 % with liver resection, and 17 % with liver transplantation [26]. In symptomatic PLD patients, mortality after liver transplantation ranged from 0 to 50 % [25]. In a Mayo Clinic cohort with age and liver-volume similar to ours—though with a smaller proportion of endstage renal disease (19 % for Mayo vs. our 56 %), 20 % died after surgical treatment, underscoring the difficulty
J Nephrol
of improving their poor condition, and perhaps suggesting that, given the poor condition of this class of patients, hepatic TAE might be a therapeutic option for patients with symptomatic PLD. To summarize, our analyses suggest that renal TAE could overcome the survival disadvantage of severe PKD. The survival benefit of hepatic TAE, however, is still unclear, though TAE may improve malnutrition caused by severe PLD and make those patients’ survival comparable to that of the general PKD population. Renal TAE could be the first-line therapeutic option for enlarged polycystic kidneys if the patient is indicated for the treatment. Acknowledgments This work was supported by JH’s research Grants from the Okinaka Memorial Institute, Toranomon Hospital, and The Kidney Foundation, Japan (JKFB13-9), by TS’s Grant from the Japanese Association of Dialysis Physicians (JADP Grant 2013-11), and by YU’s grant from the Okinaka Memorial Institute. Conflict of interest
None.
References 1. Higashihara E, Nutahara K, Kojima M et al (1998) Prevalence and renal prognosis of diagnosed autosomal dominant polycystic kidney disease in Japan. Nephron 80:421–427 2. Davies F, Coles GA, Harper PS, Williams AJ, Evans C, Cochlin D (1991) Polycystic kidney disease re-evaluated: a populationbased study. Q J Med 79:477–485 3. Torres VE, Harris PC (2009) Autosomal dominant polycystic kidney disease: the last 3 years. Kidney Int 76:149–168 4. Bae KT, Zhu F, Chapman AB et al (2006) Magnetic resonance imaging evaluation of hepatic cysts in early autosomal-dominant polycystic kidney disease: the Consortium for Radiologic Imaging Studies of Polycystic Kidney Disease cohort. Clin J Am Soc Nephrol 1:64–69 5. Bistritz L, Tamboli C, Bigam D, Bain VG (2005) Polycystic liver disease: experience at a teaching hospital. Am J Gastroenterol 100:2212–2217 6. Milutinovic J, Fialkow PJ, Rudd TG, Agodoa LY, Phillips LA, Bryant JI (1980) Liver cysts in patients with autosomal dominant polycystic kidney disease. Am J Med 68:741–744 7. Grantham JJ, Torres VE, Chapman AB et al (2006) Volume progression in polycystic kidney disease. N Engl J Med 354:2122–2130 8. Kistler AD, Poster D, Krauer F et al (2009) Increases in kidney volume in autosomal dominant polycystic kidney disease can be detected within 6 months. Kidney Int 75:235–241 9. Temmerman F, Missiaen L, Bammens B et al (2011) Systematic review: the pathophysiology and management of polycystic liver disease. Aliment Pharmacol Ther 34:702–713
10. Torres VE, Chapman AB, Devuyst O et al (2012) Tolvaptan in patients with autosomal dominant polycystic kidney disease. N Engl J Med 367:2407–2418 11. Ubara Y, Katori H, Tagami T et al (1999) Transcatheter renal arterial embolization therapy on a patient with polycystic kidney disease on hemodialysis. Am J Kidney Dis 34:926–931 12. Ubara Y, Tagami T, Sawa N et al (2002) Renal contraction therapy for enlarged polycystic kidneys by transcatheter arterial embolization in hemodialysis patients. Am J Kidney Dis 39:571–579 13. Ubara Y, Takei R, Hoshino J et al (2004) Intravascular embolization therapy in a patient with an enlarged polycystic liver. Am J Kidney Dis 43:733–738 14. Takei R, Ubara Y, Hoshino J et al (2007) Percutaneous transcatheter hepatic artery embolization for liver cysts in autosomal dominant polycystic kidney disease. Am J Kidney Dis 49:744–752 15. Hoshino J, Ubara Y, Suwabe T et al (2014) Intravascular embolization therapy in patients With enlarged polycystic liver. Am J Kidney Dis 63:937 16. Torres VE (2007) Treatment of polycystic liver disease: one size does not fit all. Am J Kidney Dis 49:725–728 17. Pei Y, Obaji J, Dupuis A et al (2009) Unified criteria for ultrasonographic diagnosis of ADPKD. J Am Soc Nephrol 20:205–212 18. Patient_Registration_Committee (2011) An overview of regular dialysis treatment in Japan (as of 31 December 2011), CD-ROM edition. Japanese Society for Dialysis Therapy, Tokyo 19. Nakai S, Iseki K, Itami N et al (2012) Overview of regular dialysis treatment in Japan (as of 31 December 2009). Ther Apher Dial 16:11–53 20. Higashihara E, Horie S, Muto S, Mochizuki T, Nishio S, Nutahara K (2012) Renal disease progression in autosomal dominant polycystic kidney disease. Clin Exp Nephrol 16:622–628 21. Fick GM, Johnson AM, Hammond WS, Gabow PA (1995) Causes of death in autosomal dominant polycystic kidney disease. J Am Soc Nephrol 5:2048–2056 22. Suwabe T, Ubara Y, Sumida K et al (2012) Clinical features of cyst infection and hemorrhage in ADPKD: new diagnostic criteria. Clin Exp Nephrol 16:892 23. Sherstha R, McKinley C, Russ P et al (1997) Postmenopausal estrogen therapy selectively stimulates hepatic enlargement in women with autosomal dominant polycystic kidney disease. Hepatology 26:1282–1286 24. Park HC, Kim CW, Ro H et al (2009) Transcatheter arterial embolization therapy for a massive polycystic liver in autosomal dominant polycystic kidney disease patients. J Korean Med Sci 24:57–61 25. Abu-Wasel B, Walsh C, Keough V, Molinari M (2013) Pathophysiology, epidemiology, classification and treatment options for polycystic liver diseases. World J Gastroenterol 19:5775– 5786 26. Drenth JP, Chrispijn M, Nagorney DM, Kamath PS, Torres VE (2010) Medical and surgical treatment options for polycystic liver disease. Hepatology 52:2223–2230
123
ORIGINAL ARTICLE
Survival after arterial embolization therapy in patients with polycystic kidney and liver disease Junichi Hoshino • Tatsuya Suwabe • Noriko Hayami • Keiichi Sumida Koki Mise • Masahiro Kawada • Aya Imafuku • Rikako Hiramatsu • Masayuki Yamanouchi • Eiko Hasegawa • Naoki Sawa • Ryoji Takei • Kenmei Takaichi • Yoshifumi Ubara
•
Received: 22 May 2014 / Accepted: 1 September 2014 Ó Italian Society of Nephrology 2014
Abstract Background Transcatheter arterial embolization (TAE) has become a therapeutic option for symptomatic polycystic kidney disease (PKD) and polycystic liver disease (PLD). However, factors affecting survival with renal TAE remain unknown. Methods All symptomatic patients with severe PKD and/ or PLD who received renal and/or hepatic TAE at our center from October 1996 through March 2013 (n = 1,028) were followed until death. Their survival was compared with that of the general PKD population on dialysis in Japan. Factors affecting survival were analyzed using the Cox hazard model. Results After renal TAE, 5- and 10-year survival was, respectively, 0.78 (95 % confidence interval, 0.74–0.82) and 0.56 (0.49–0.63); with hepatic TAE, 0.69 (0.58–0.77) and 0.41 (0.22–0.60); and with dual TAE (renal and hepatic), 0.82 (0.72–0.88) and 0.45 (0.31–0.59). Survival after dialysis initiation was better among patients with renal TAE than among general PKD patients. Factors Electronic supplementary material The online version of this article (doi:10.1007/s40620-014-0138-0) contains supplementary material, which is available to authorized users. J. Hoshino (&) T. Suwabe N. Hayami K. Sumida K. Mise K. Takaichi Y. Ubara Nephrology Center, Toranomon Hospital, 1-3-1, Kajigaya, Takatsu-ku, Kawasaki, Kanagawa 213-8587, Japan e-mail: [email protected] J. Hoshino T. Suwabe N. Hayami K. Sumida M. Kawada A. Imafuku R. Hiramatsu M. Yamanouchi E. Hasegawa N. Sawa K. Takaichi Y. Ubara Nephrology Center, Toranomon Hospital, Tokyo, Japan R. Takei Department of Radiology, Touzan Hospital, Tokyo, Japan
affecting survival after renal TAE were age [hazard ratio (HR) 3.02 (1.44–6.33) for every 10 years] and albumin [HR 0.70 (0.55–0.89) per 0.1 g/dl]. Kidney volume was not associated with patient death after TAE. The main causes of death among patients after renal TAE were similar to those of the general PKD population on dialysis whereas, after hepatic TAE, the main cause was cyst infection with liver failure (12.5 % with PLD and 5.9 % with PKD, p \ 0.01). Conclusion Survival after renal TAE with severe PKD was better than for the general PKD population on dialysis, suggesting that renal TAE could overcome the disadvantage due to huge organ size. Keywords Autosomal dominant polycystic kidney disease (ADPKD) Hepatic transcatheter arterial embolization (hepatic TAE) Renal transcatheter arterial embolization (renal TAE) Polycystic liver disease (PLD) Survival Cause of death
Introduction Autosomal dominant polycystic kidney disease (ADPKD) is the most common hereditary kidney disease [1, 2] and polycystic liver disease (PLD) its most common extrarenal manifestation [3]. Several studies reported that over 80 % of ADPKD patients had hepatic cysts [4, 5], risk factors for developing them being severity of renal cysts and poor renal function [5, 6]. Huge kidneys and livers may lead to malnutrition and can be lethal, with kidney and cyst volumes reportedly increasing exponentially, annual growth rate calculated at 5.27–5.36 % [7, 8]. Conservative treatment approaches are only partially effective and do not change the disease’s course [9, 10]. Accordingly, in
123
J Nephrol
October 1996 we initiated transcatheter arterial embolization (TAE) for treating symptomatic severe ADPKD/PLD patients (i.e., PKD and/or PLD with huge organ size), and noted mass volume reduction and nutritional improvement in both kidney and liver patients [11–15]. After renal TAE, over 50 % reduction in kidney volume and significant improvement in anemia and nutritional status were observed. With hepatic TAE, though—while mild reduction of cystic volume reduction and improvement of anemia were observed—the results were not as dramatic because the hepatic TAE embolized only cyst-occupied segments and hepatic parenchymal improvement was also seen with hepatic cyst contraction [15]. Hepatic TAE has become an accepted treatment for highly symptomatic patients who have large areas of cystic liver without recognizable hepatic parenchyma, and are not good candidates for combined liver resection, cyst fenestration, or liver transplantation [3, 13, 14, 16]. In the present longitudinal study, we aimed to clarify the rate of survival after renal and hepatic TAE, compare causes of death, and analyze the effectiveness of these treatments by comparing the survival with that reflected in the national data for Japanese dialysis patients (the general PKD/PLD population, some of whom may have received TAE unknown to us).
Subjects and methods Patients and study design From October 1996 through March 2013, 1,028 ADPKD patients received TAE treatment in our facilities. Diagnosis of polycystic kidney and liver was established by computed tomography (CT), magnetic resonance imaging, and ultrasonography with Ravine’s revised unified diagnostic criteria [17]. Inclusion criteria for renal TAE were: being on maintenance dialysis with daily urinary volume less than 500 ml/day; not being a good candidate for percutaneous cyst aspiration with/without sclerosis, or for surgery (or having declined other interventions); and having compression symptoms related to enlarged polycystic kidney or active renal cyst bleeding. Patients with uncontrollable renal cyst infection were excluded. Inclusion criteria for hepatic TAE—described previously [15]—were: having symptomatic PLD due to mass effects from hepatic segments almost completely occupied by cysts (over 90 % per segment), but with at least one segment almost intact, maintaining functioning liver parenchyma; and not being a candidate for surgery (or having declined other interventions). Exclusion criteria for hepatic TAE were: severe liver dysfunction (total bilirubin [2.0 mg/dl), massive ascites, diffusely
123
distributed intrahepatic cysts, or severe cyst infection associated with malnutrition. All TAE patients had expressed a strong desire for TAE therapy and gave consent after being fully informed about TAE’s risks. Of the 1,028 ADPKD patients, 68 % (n = 695, 45 % female) who had a markedly enlarged polycystic kidney received renal TAE, 21 % (n = 221, 77 % female) with marked PLD without a huge polycystic kidney received hepatic TAE, and 11 % (n = 112, 73 % female) having both an enlarged kidney and liver received dual TAE. In all— because of dual treatments—807 patients received renal TAE, and 333 received hepatic TAE. Looking at initial treatment (even if it was the only TAE received), 762 patients received renal TAE and 266 received hepatic TAE. With dual TAE, 67 of 112 (59.8 %) received renal TAE first. Interestingly, over 80 % [81 % in chronic kidney disease (CKD) G4-5, 85 % in CKD G1-3] of PLD patients with preserved renal function were female, suggesting that gender strongly affects PLD development. Patients’ demographic data were obtained at the time of TAE, cause of death from medical records. All laboratory data—including serum total protein (TP), serum albumin (Alb), serum creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), cholinesterase (ChE), c-glutamyl transpeptidase (GGT), C-reactive protein (CRP), hemoglobin (Hb), abdominal circumference, cardiac ejection fraction, and kidney/liver volume—were collected before treatment and 3, 6, and 12 months afterwards. Measurement of organ volume was described previously [13–15]. Organ volumes were measured from the set of contiguous images by summing the products of the organ areas traced in each CT image and the slice thickness. All laboratory values were measured by the automated, standardized methods used in our hospital within 24 h after drawing blood samples. The primary outcome of this study was death, with all patients followed until death or end of follow-up, using medical records, telephoning and/or writing to them. Methods of renal and hepatic TAE Methods of administering renal and hepatic TAE have been detailed previously [11–15]. To summarize, we performed renal and hepatic angiography using Seldinger’s technique, with a five-French shepherd hook catheter (Cathex, Shibuya, Tokyo, Japan) for renal angiography, and, for hepatic angiography, two catheters together (2- and 4-French) to navigate the hairpin-like meander of hepatic arterial branches. After the targeted arterial branches had been evaluated and selected, the arteries were carefully embolized from the periphery using platinum microcoils with a 0.46 mm diameter and 20–140 mm length (C-stopper coil, Piolax Medical Devices Inc., Yokohama, Japan; and
J Nephrol
TornadoÒ Embolization Coils, Cook Medical Inc., Bloomington, IN, USA). Usually, patients had fever and abdominal pain for a week, controllable with pentazocine, hydroxyzine hydrochloride, and nonsteroidal anti-inflammatory drugs. In renal TAE, an epidural catheter was inserted into the tenth or eleventh thoracic vertebra to control the severer pain of kidney embolization. Complications attending the procedure were pseudoaneurysm of the femoral artery (0.6 %) and the microcoil entering the aorta (0.8 %). Serious complications after renal TAE were emphysematous cyst infection (0.2 %) and disseminated intravascular coagulation (0.1 %). No serious treatment-related adverse effects were observed after hepatic TAE [15].
rank test. To eliminate the immortal person-time effect, we excluded patients who died within 2–4 years. The Cox proportional hazard model was used to obtain hazard ratios (HRs) and a 95 % confidence interval (CI) for death after TAE adjusted for sex, age, dialysis vintage, Alb, ALP, CRP, and kidney volume. We categorized patients into quartiles by original organ volume after treatment, dual TAE recipients divided by initial organ volume. All analyses used StataÒ SE version 12.0 (StataCorp, College Station, TX, USA).
Results Patients’ characteristics
Statistical analysis Data were summarized using proportions and means [±standard deviation (SD)] as appropriate. Categorical variables were analyzed with the Chi-squared or Fisher’s exact test, continuous variables compared using Student’s t test, Mann–Whitney U test, or analysis of variance (ANOVA). Cumulative survival was estimated with Kaplan–Meier survival curves, and compared using the logTable 1 Characteristics of ADPKD patients who received TAE
Table 1 and Fig. 1 show the breakdown and characteristics of ADPKD patients who received TAE. With hepatic TAE, mean TP, Alb, AST, ALT, ALP, GGT, and Hb were higher than in the other two groups (p \ 0.01) and mean age was lower (53.4 ± 9.2 years for hepatic TAE vs. 57.9 ± 9.0 for renal and 56.3 ± 6.8 for dual TAE, p \ 0.01). Median organ volumes were similar in all groups: median kidney volumes with renal and dual TAE were 4,667 and
Total (n = 1,028)
Renal TAE (n = 695)
Dual TAE (n = 112)
Hepatic TAE (n = 221)
p value
Sex (female %)
55.1 %
45.2 %
73.2 %
76.8 %
\0.01
Age (years)
56.7 ± 9.0
57.9 ± 9.0
56.3 ± 6.8
53.4 ± 9.2
\0.01 \0.01
Under dialysis (%)
85.3 %
100.0 %
96.4 %
33.5 %
Dialysis vintage (years)
5.8 ± 5.2
6.2 ± 5.3
3.9 ± 4.1
5.7 ± 5.1
0.02
Total protein (g/dl)
6.8 ± 0.6
6.7 ± 0.6
6.8 ± 1.0
7.0 ± 0.6
\0.01 \0.01
Albumin (g/dl)
3.1 ± 0.4
3.1 ± 0.4
3.1 ± 0.4
3.3 ± 0.4
AST (IU/l)
18 ± 16
16 ± 17
17 ± 7
23 ± 16
\0.01
ALT (IU/l)
13 ± 15
11 ± 13
12 ± 6
18 ± 19
\0.01 \0.01
ALP (IU/l)
287 ± 227
261 ± 220
275 ± 128
361 ± 250
ChE (IU/l)
174 ± 98
179 ± 95
167 ± 97
164 ± 105
0.16
GGT (IU/l) CRP (mg/dl)
81 ± 108 1.0 ± 2.8
57 ± 78 1.1 ± 3.2
86 ± 56 1.0 ± 2.9
142 ± 155 0.7 ± 1.1
\0.01 0.19
Hemoblobin (g/dl)
10.4 ± 1.6
10.2 ± 1.5
9.9 ± 1.2
11.0 ± 1.8
\0.01
6,433 (4,6389,463)
5,656 (4,3447,723)
0.3
Target organ volume Liver volume (cm3) Kidney volume (cm3) ADPKD autosomal dominant polycystic kidney disease, TAE transcatheter arterial embolization, AST aspartate aminotransferase, ALT alanine aminotransferase, ALP alkaline phosphatase, ChE cholinesterase, GGT c-glutamyl transpeptidase, CRP C-reactive protein
4,667 (3,2896,471)
5,468 (3,8806,256)
0.45
Abdominal circumference (cm)
92.1 ± 12.3
93.9 ± 10.4
89.8 ± 6.4
91.5 ± 13.4
0.4
Cardiac ejection fraction (%)
68 ± 10
67 ± 11
70 ± 9
71 ± 8
0.03
Mean TAE duration (year)
1.9 ± 2.1 2.0 ± 1.9 (liver first) (n = 45) 1.9 ± 2.2 (kidney first) (n = 67)
123
J Nephrol Fig. 1 Breakdown of ADPKD patients treated with TAE ADPKD autosomal dominant polycystic kidney disease; TAE transcatheter arterial embolization. Chronic kidney disease (CKD) G1-3: estimated glomerular filtration rate (eGFR) C30 ml/min/1.73 m2; G4, 5: eGFR \30 ml/min/ 1.73 m2
5,468 cm3, respectively (p = 0.45), median liver volumes with hepatic and dual TAE were 5,656 and 6,433 cm3, respectively (p = 0.30). Mean abdominal circumferences were also similar among the three groups (p = 0.40). The mean interval between sequential treatments with dual TAE was 1.9 ± 2.1 years (2.0 ± 1.9 with hepatic TAE first, and 1.9 ± 2.2 years with renal TAE first). Survival after TAE Comparing post-TAE survival curves, overall survival among the three groups was similar (p = 0.98), though slightly better with only renal TAE (Fig. 2a). With renal TAE, 3-, 5-, and 10-year survival was, respectively: 0.87 (95 % CI, 0.84–0.90), 0.78 (0.74–0.82), and 0.56 (0.49–0.63); with hepatic TAE, 0.86 (0.80–0.91), 0.69 (0.58–0.77), and 0.41 (0.22–0.60); and with dual TAE, 0.94 (0.88–0.97), 0.82 (0.72–0.88), and 0.45 (0.31–0.59). Next, we compared survival after renal TAE among patients with and without sequential hepatic TAE to determine the effect of hepatic TAE. For dual TAE patients who had renal treatment first, 3-, 5-, and 10-year survival was, respectively, 0.95 (0.85–0.98), 0.81 (0.68–0.90), and 0.42 (0.24–0.59) (Fig. 2b). That curve did not significantly differ from the curve for renal TAE (p = 0.29, log-rank). However, because mean duration between renal and hepatic TAE with dual treatment was 1.9 ± 2.2 years, one must exclude those surviving B2 years in order to minimize the effect of the duration of sequential TAE treatment
123
(i.e., immortal person-time). Considering that exclusion, survival with renal TAE was significantly better than with dual, kidney-first TAE (p = 0.03, log-rank), suggesting that the prognosis of patients with PLD was worse—even after hepatic TAE—than for patients who were nonsymptomatic or had no PLD (Fig. 2c). Then we compared survival after hepatic TAE among patients with and without sequential renal TAE to determine the effect of sequential renal TAE on those with huge kidneys. After dual TAE with hepatic TAE first, 3-, 5-, and 10-year survival was, respectively, 0.93 (0.80–0.98), 0.82 (0.64–0.91), and 0.51 (0.25–0.71) (Fig. 2d). Although overall survival after hepatic TAE without renal TAE seemed inferior to that with renal TAE (p = 0.23, logrank), again, it was similar between the two groups for those surviving [2 years (p = 0.83, log-rank, Fig. 2e). Considering the immortal person-time effect, the prognosis with severe PKD after renal TAE was very close to that for non-symptomatic PKD patients. We analyzed factors affecting survival after renal TAE with Cox models (Table 2). In the adjusted model, HRs Fig. 2 Overall survival after TAE a overall survival among patients c who received renal transcatheter arterial embolization (TAE), hepatic TAE, or dual TAE (sequential renal and hepatic TAE). b Overall survival after renal TAE with or without hepatic TAE. The mean duration of renal TAE and hepatic TAE with dual TAE was 1.9 ± 2.2 years. c Survival after renal TAE for those surviving more than 2 years. d Overall survival after hepatic TAE with or without renal TAE. The mean duration of hepatic TAE and renal TAE with dual TAE was 2.0 ± 1.9 years
J Nephrol
123
J Nephrol Table 2 Hazard ratios of all causes of mortality after renal TAE
Model 1 (unadjusted) HR
95 % CI-L
95 % CI-H
Model 2 (adjusted) p value
HR
95 % CI-L
95 % CI-H
p value
0.12
Kidney volume (cm3)
Model was adjusted for TAE type, sex, age, Alb, ALP, CRP, dialysis vintage, and target organ volume TAE transcatheter arterial embolization, Alb serum albumin, ALP alkaline phosphatase, CRP C-reactive protein, HR hazard ratio, CI confidence interval
Quartile 1 (mild)
1.00
Quartile 2
0.79
0.35
1.79
0.57
0.24
0.04
1.44
Quartile 3
0.36
0.13
1.00
0.50
0.63
0.12
3.36
0.60
Quartile 4 (severe)
1.04
0.50
2.15
0.93
1.26
0.27
5.87
0.77
Male sex
1.41
1.05
1.91
0.02
0.50
0.15
1.67
0.26
Dialysis vintage B5 years
1.00
1.00
5–10 years
0.99
0.62
1.60
0.98
0.64
0.14
2.87
0.56
[10 years
2.19
1.38
3.48
0.001
0.46
0.10
2.03
0.31
Age (every 10 years older)
1.89
1.56
2.29
\0.001
3.02
1.44
6.33
0.003
Alb (per 0.1 g/dl)
0.87
0.83
0.91
\0.001
0.70
0.55
0.89
0.004 0.36
ALP (per 100 IU/l)
1.20
1.15
1.25
\0.001
1.09
0.91
1.30
CRP (mg/dl)
1.04
1.01
1.07
0.01
1.03
0.85
1.24
0.78
Dual treatment with hepatic TAE
1.13
0.79
1.61
0.51
0.15
0.01
2.72
0.20
among kidney-volume groups were not significant, though the HR in patients with the largest organ volume—compared to that of those with the smallest volume—was 1.26 (0.27–5.87). Age [HR 3.02 (1.44–6.33) for every 10 years of patient age] and Alb [HR 0.70 (0.55–0.89) per 0.1 g/dl] were significant survival factors. Survival after initiation of dialysis with/without TAE The next question was whether TAE improved survival for its recipients compared to survival of the general PKD population on dialysis in Japan as reported annually by the Japanese Society of Dialysis Therapy (JSDT) [18]. According to that report, average 3-, 5-, 10- and 15-year survival for the Japanese PKD population on hemodialysis (henceforth, ‘‘JSDT population’’) was, respectively, 0.86, 0.77, 0.58, and 0.41. In comparison, survival after initiation of dialysis was significantly better with renal TAE than for the other two TAE/dialysis groups and the JSDT population (p \ 0.01, log-rank), but survival was similar with hepatic-only and dual TAE, and for the JSDT population. For instance, with renal TAE, 5-year survival was 0.97 (0.94–0.98) vs. the JSDT population’s 0.77, whereas with only hepatic TAE it was 0.78 (0.66–0.86) and with dual TAE, 0.85 (0.76–0.91). Figure 3 charts each time point. Since mean dialysis vintage at the time of TAE was 6.2 ± 5.3 years with renal TAE, 5.7 ± 5.1 years with hepatic TAE, and 3.9 ± 4.1 years with dual TAE, we next calculated patients’ survival after initiation of dialysis with 6-year censoring for renal and hepatic TAE, but 4-year
123
1.00
censoring for dual TAE group to minimize the immortal person-time effect. Again, survival with renal TAE was significantly better than for the JSDT population (Fig. 3a), though survival with hepatic TAE was a little worse at 10 years and only a bit better at 15 years. These findings suggested that renal TAE could overcome severe PKD’s disadvantage of malnutrition, whereas hepatic TAE’s effect for severe PLD was equivocal (Fig. 3b). In addition, with 4-year censoring, survival for dual TAE after initiation of dialysis was very similar to that of the general population (Fig. 3c). Causes of death Table 3 shows causes of death for TAE patients. Interestingly, the causes of death with renal and hepatic TAE were quite different. With renal TAE, the main causes were cardiovascular (3.9 % of all the patients in the treatment group) and cerebrovascular (2.4 %), much as with the JSDT population [18], while the death rate from liver failure resulting from liver cyst infection was significantly higher with hepatic TAE (5.9 %) and dual TAE (12.5 %) than with renal TAE (1.7 %) (p \ 0.01); comparison of dual TAE’s and hepatic TAE’s death rates from liver failure (5.9 % vs. 12.5 %) was p = 0.054 by Fisher’s exact test. The rate of other infectious deaths was slightly higher for dual TAE (2.7 %) than for renal (1.0 %) and hepatic (9 %) (p = 0.15, Fisher’s exact test). Note that the proportion of death with dual TAE was significantly higher than in the other two groups (34.8 vs. 20.3 % renal and
J Nephrol Fig. 3 Overall survival after initiation of dialysis. a overall survival among patients on dialysis who received renal transcatheter arterial embolization (TAE), hepatic TAE, or dual TAE (sequential renal and hepatic TAE). Survival data for the Japanese PKD population on dialysis is from the annual report (as of 31 December 2011) published by the Japanese Society for Dialysis Therapy. *It was p \ 0.01 between renal TAE and other groups including the general PKD/PLD population. It was p = 0.07 between the dual TAE and hepatic TAE groups. b and c Survival after initiation of dialysis for those surviving more than 4 and 6 years. The mean dialysis vintage at the time of TAE was 6.2 ± 5.3 years with renal TAE, 5.7 ± 5.1 years with hepatic TAE, and 3.9 ± 4.1 years with dual TAE
19.5 % hepatic, p \ 0.01), suggesting poorer prognosis in patients with severe PKD and PLD.
Discussion We studied the survival of Japanese patients with severe PKD and/or PLD after TAE. In Japan, ADPKD is the fourth leading cause of primary disease for patients on maintenance dialysis (3.4 %) [19]. Yet treatment options for severe symptomatic PKD and PLD remain very limited. Previously, we reported organ volume reduction and nutritional improvement with renal and hepatic TAE [11– 15]. As a result, more than 1,000 patients have come to our
hospital wanting TAE. Their median kidney volume was 4,667 cm3, over two SD larger than the mean kidney volume reported in Japan [20]—1,839 ± 1,329 cm3—suggesting the poorer general condition of our cohort. Based on our analyses, five things seem certain. First, 5and 10-year survival of patients after renal TAE was, respectively, 78 and 56 %, and after hepatic TAE 69 and 41 %. Second, outcome for patients after renal TAE was better than for patients without renal TAE—even with the immortal person-time effect—suggesting that renal TAE can overcome the disadvantage of enlarged kidneys. Third, the effect of hepatic TAE may be limited: in our analyses, survival after hepatic TAE was inferior to that of patients without symptomatic PLD; so hepatic TAE may not be
123
J Nephrol Table 3 Causes of death after TAE
Renal
TAE
Hepatic
TAE
221
Dual
TAE
112
Total
Number of patients
695
Total deaths
141
20.3 %
43
19.5 %
39
34.8 %a,b
1,028 223
Cyst infection/liver failure
12
1.7 %
13
5.9 %a
14
12.5 %a
39
3.8 %
Infection (pneumonia etc.)
7
1.0 %
2
0.9 %
3
2.7 %
12
1.2 %
21.7 % 0.0 %
The number of patients who died of each cause is shown, as well as the percentage of total cohort deaths each number represents
Malignancy
12
1.7 %
2
0.9 %
1
0.9 %
15
1.5 %
Cardiovascular
27
3.9 %
4
1.8 %
3
2.7 %
34
3.3 %
Cerebrovascular
17
2.4 %
4
1.8 %
3
2.7 %
24
2.3 %
Respiratory failure
1
0.1 %
0
0.0 %
0
0.0 %
1
0.1 %
a
Sudden death
9
1.3 %
0
0.0 %
1
0.9 %
10
1.0 %
p \ 0.01, comparing to renal TAE b
p \ 0.01, comparing to hepatic TAE
Other (accident etc.)
17
2.4 %
5
2.3 %
6
5.4 %
28
2.7 %
Unknown
39
5.6 %
13
5.9 %
8
7.1 %
60
5.8 %
enough to overcome the survival disadvantage of severe PLD. Fourth, the trend in causes of death differed between patients with renal and hepatic TAE. The main cause of death with severe PLD was liver cystic infection leading to liver failure, whereas for patients with huge kidneys the main causes were cardiovascular and cerebrovascular disease, the same as for the JSDT population. Since infections are reportedly the second most common cause of ADPKD patient death [21]—and most cyst infection occurs in livers [22]—the difference in causes of death with hepatic TAE may be reasonable. Nonetheless, the low percentage of infectious death with renal TAE seems notable—perhaps partly attributable to the marked nutritional improvement after renal TAE. Finally, although gender distribution of ADPKD patients with enlarged kidneys was almost equal, over 80 % of ADPKD patients with an enlarged liver that had preserved renal function were female, suggesting a different mechanism of cystic growth for renal and hepatic cysts: estrogen reportedly affects hepatic cyst growth, but not renal cystic growth [23]. Perhaps female hormonal change decreases the gender difference in prevalence of severe PLD patients on dialysis. There are some limitations to this study. First, referral bias may limit the external validity of our results. However, according to a hospital-based study of ADPKD prevalence in Japan [1], our cohort appears representative of Japanese ADPKD patients with severe PKD and PLD. Moreover, the mean age of our population and of those in that report was similar (56.7 ± 9.0 and 53.3 ± 15.9 years), so the problem of external validity may be insignificant. Second, this study did not have control patients—and could not—because all patients had severe conditions and were contraindicated for other therapies or were referred to our hospital because of a strong desire to receive TAE. Comparing survival with renal or hepatic TAE versus that with dual TAE (or of the JSDT population) could minimize this limitation. Third, our patients were younger than the JSDT population (their
123
mean age, 56.7 ± 9.0 vs. 64.4 ± 11.2 years). However, survival curve trends were very similar to our initial findings if we compared only our age-matched patients (64.3 ± 4.9 years) with the JSDT population (supplementary Fig. 1). Fourth, improvement of treatment technologies may confound results in long-term follow-up studies. However, survival after renal TAE in our study was similar in different periods: 5-year survival was 77.4 % in 1996–2000, 80.1 % in 2001–2005, and 73.7 % 2006–2008, implying that treatment year may not have had a large impact on our analyses. Finally, more experiences with selective renal/hepatic artery embolization by other groups and/or randomized prospective studies are needed to confirm the safety and efficacy of this procedure. Although TAE is increasingly recognized in Japan and Korea as a non-invasive option for highly symptomatic PKD/PLD patients who are not good candidates for surgical treatment [11–14, 24], TAE in lieu of surgery is not yet established as part of the medical culture in Western countries. The reason for this difference could be the higher rate of kidney/liver transplantation in the West, with cystic kidneys removed at the time of transplantation. In our study, 5-year mortality after dual TAE was 15 %, comparable to the expected 20 % 5-year mortality for combined liver/kidney transplantation in PLD patients [25]. In addition, 5-year mortality after hepatic TAE was 22 % in our study. It was reported that for PLD patients in better general condition than ours, overall mortality of traditional invasive approaches was 2 % with fenestration, 3 % with liver resection, and 17 % with liver transplantation [26]. In symptomatic PLD patients, mortality after liver transplantation ranged from 0 to 50 % [25]. In a Mayo Clinic cohort with age and liver-volume similar to ours—though with a smaller proportion of endstage renal disease (19 % for Mayo vs. our 56 %), 20 % died after surgical treatment, underscoring the difficulty
J Nephrol
of improving their poor condition, and perhaps suggesting that, given the poor condition of this class of patients, hepatic TAE might be a therapeutic option for patients with symptomatic PLD. To summarize, our analyses suggest that renal TAE could overcome the survival disadvantage of severe PKD. The survival benefit of hepatic TAE, however, is still unclear, though TAE may improve malnutrition caused by severe PLD and make those patients’ survival comparable to that of the general PKD population. Renal TAE could be the first-line therapeutic option for enlarged polycystic kidneys if the patient is indicated for the treatment. Acknowledgments This work was supported by JH’s research Grants from the Okinaka Memorial Institute, Toranomon Hospital, and The Kidney Foundation, Japan (JKFB13-9), by TS’s Grant from the Japanese Association of Dialysis Physicians (JADP Grant 2013-11), and by YU’s grant from the Okinaka Memorial Institute. Conflict of interest
None.
References 1. Higashihara E, Nutahara K, Kojima M et al (1998) Prevalence and renal prognosis of diagnosed autosomal dominant polycystic kidney disease in Japan. Nephron 80:421–427 2. Davies F, Coles GA, Harper PS, Williams AJ, Evans C, Cochlin D (1991) Polycystic kidney disease re-evaluated: a populationbased study. Q J Med 79:477–485 3. Torres VE, Harris PC (2009) Autosomal dominant polycystic kidney disease: the last 3 years. Kidney Int 76:149–168 4. Bae KT, Zhu F, Chapman AB et al (2006) Magnetic resonance imaging evaluation of hepatic cysts in early autosomal-dominant polycystic kidney disease: the Consortium for Radiologic Imaging Studies of Polycystic Kidney Disease cohort. Clin J Am Soc Nephrol 1:64–69 5. Bistritz L, Tamboli C, Bigam D, Bain VG (2005) Polycystic liver disease: experience at a teaching hospital. Am J Gastroenterol 100:2212–2217 6. Milutinovic J, Fialkow PJ, Rudd TG, Agodoa LY, Phillips LA, Bryant JI (1980) Liver cysts in patients with autosomal dominant polycystic kidney disease. Am J Med 68:741–744 7. Grantham JJ, Torres VE, Chapman AB et al (2006) Volume progression in polycystic kidney disease. N Engl J Med 354:2122–2130 8. Kistler AD, Poster D, Krauer F et al (2009) Increases in kidney volume in autosomal dominant polycystic kidney disease can be detected within 6 months. Kidney Int 75:235–241 9. Temmerman F, Missiaen L, Bammens B et al (2011) Systematic review: the pathophysiology and management of polycystic liver disease. Aliment Pharmacol Ther 34:702–713
10. Torres VE, Chapman AB, Devuyst O et al (2012) Tolvaptan in patients with autosomal dominant polycystic kidney disease. N Engl J Med 367:2407–2418 11. Ubara Y, Katori H, Tagami T et al (1999) Transcatheter renal arterial embolization therapy on a patient with polycystic kidney disease on hemodialysis. Am J Kidney Dis 34:926–931 12. Ubara Y, Tagami T, Sawa N et al (2002) Renal contraction therapy for enlarged polycystic kidneys by transcatheter arterial embolization in hemodialysis patients. Am J Kidney Dis 39:571–579 13. Ubara Y, Takei R, Hoshino J et al (2004) Intravascular embolization therapy in a patient with an enlarged polycystic liver. Am J Kidney Dis 43:733–738 14. Takei R, Ubara Y, Hoshino J et al (2007) Percutaneous transcatheter hepatic artery embolization for liver cysts in autosomal dominant polycystic kidney disease. Am J Kidney Dis 49:744–752 15. Hoshino J, Ubara Y, Suwabe T et al (2014) Intravascular embolization therapy in patients With enlarged polycystic liver. Am J Kidney Dis 63:937 16. Torres VE (2007) Treatment of polycystic liver disease: one size does not fit all. Am J Kidney Dis 49:725–728 17. Pei Y, Obaji J, Dupuis A et al (2009) Unified criteria for ultrasonographic diagnosis of ADPKD. J Am Soc Nephrol 20:205–212 18. Patient_Registration_Committee (2011) An overview of regular dialysis treatment in Japan (as of 31 December 2011), CD-ROM edition. Japanese Society for Dialysis Therapy, Tokyo 19. Nakai S, Iseki K, Itami N et al (2012) Overview of regular dialysis treatment in Japan (as of 31 December 2009). Ther Apher Dial 16:11–53 20. Higashihara E, Horie S, Muto S, Mochizuki T, Nishio S, Nutahara K (2012) Renal disease progression in autosomal dominant polycystic kidney disease. Clin Exp Nephrol 16:622–628 21. Fick GM, Johnson AM, Hammond WS, Gabow PA (1995) Causes of death in autosomal dominant polycystic kidney disease. J Am Soc Nephrol 5:2048–2056 22. Suwabe T, Ubara Y, Sumida K et al (2012) Clinical features of cyst infection and hemorrhage in ADPKD: new diagnostic criteria. Clin Exp Nephrol 16:892 23. Sherstha R, McKinley C, Russ P et al (1997) Postmenopausal estrogen therapy selectively stimulates hepatic enlargement in women with autosomal dominant polycystic kidney disease. Hepatology 26:1282–1286 24. Park HC, Kim CW, Ro H et al (2009) Transcatheter arterial embolization therapy for a massive polycystic liver in autosomal dominant polycystic kidney disease patients. J Korean Med Sci 24:57–61 25. Abu-Wasel B, Walsh C, Keough V, Molinari M (2013) Pathophysiology, epidemiology, classification and treatment options for polycystic liver diseases. World J Gastroenterol 19:5775– 5786 26. Drenth JP, Chrispijn M, Nagorney DM, Kamath PS, Torres VE (2010) Medical and surgical treatment options for polycystic liver disease. Hepatology 52:2223–2230
123
