Patient, kidney, and pancreas survival in pancreas after kidney transplantation versus simultaneous pancreas and kidney transplantation: meta-analysis.

Dear Editor

Both pancreas after kidney transplant (PAK) and simultaneous pancreas and kidney transplant (SPK) are treatment options for patients with type 1 diabetes (T1DM) and end-stage kidney disease (ESKD), but they have different risks and benefits. Presently, SPK is more widespread than PAK because SPK requires only one operation[1]. PAK is a recommended option for patients with living kidney donors to avoid long-term uraemia while waiting for cadaveric pancreas transplantation and to increase the survival of allograft kidney transplantation[2]. The aim of this study was to compare PAK and SPK in terms of patient, kidney, and pancreas survival.

Online databases were used to locate studies of patient with T1DM and ESKD undergoing PAK and SPK up to 20 January 2022 ( and ). A total of 47 986 patients were identified from 24 studies, including 9093 patients undergoing PAK and 38 893 patients undergoing SPK (. Patient, kidney, pancreas survival rate as well as human leucocyte antigens (HLA) mismatch rate in patients undergoing PAK versus SPK at 1, 3, 5, and 10 years in the included studies are shown in . Quality ranged on the Newcastle–Ottawa Scale from six to nine (. The definitions of pancreas and kidney rejection, the rejection ratios of PAK and SPK, and the details of the rejection treatment used in all the included articles are given in .

Patients with PAK had a significantly higher 1-year patient survival than patients with SPK (OR 1.11, 95 per cent c.i. 1.00 to 1.24), whereas patients with PAK had a lower 10-year patient survival than patients with SPK (OR 0.73, 95 per cent c.i. 0.67 to 0.79) (). Patients with PAK had a higher 1-year (OR 5.72, 95 per cent c.i. 4.38 to 7.48), 3-year (OR 2.51, 95 per cent c.i. 2.21 to 2.84), 5-year (OR 1.74, 95 per cent c.i. 1.37 to 2.21), and 10-year (OR 1.25, 95 per cent c.i. 1.16 to 1.36) kidney graft survival than after SPK (), but lower 1-year (OR 0.79, 95 per cent c.i. 0.66 to 0.94), 3-year (OR 0.55, 95 per cent c.i. 0.46–0.65), and 5-year (OR 0.55, 95 per cent c.i. 0.44 to 0.69) pancreas graft survival than after SPK (). The incidence of pancreas rejection (OR 2.37, 95 per cent c.i. 1.47 to 3.85) and the HLA mismatch rate (OR 2.22, 95 per cent c.i. 1.19 to 4.15) were higher in patients with PAK than with SPK ( and ).

Patients with SPK have better long-term survival than patients with PAK, which may indicate that rejection is difficult to detect without simultaneous kidney transplantation[3]. Kidney graft survival in PAK is significantly better, in part due to selection bias, as only recipients with good renal function after kidney transplantation will proceed to pancreas transplantation. Another main reason for higher kidney graft survival in the PAK group was that the proportion of living donors was significantly higher. The pancreas graft long-term survival rate in PAK is still slightly lower than that of SPK, and an adverse pancreas outcome after PAK may be due to technical complications and immunological issues[4]. Higher graft survival yields lower mortality and preventing both immune and non-immune causes of graft failure is critical in reducing post-transplant mortality[5].

This study has some limitations, the studies included a large range of years, and earlier transplantation technology had a certain impact on the patient survival and graft survival of the two groups (SPK versus PAK recipients had a higher incidence of kidney graft loss due to technical reasons in 2000: 2.1 per cent versus 0 per cent . At present, the technical failure rate of SPK as a cause of allograft failure is similar to that of PAK). Furthermore, limited to the results in the included literature, the effects of other post-transplant complications, such as thrombosis, infection, obesity, smoking, and coronary heart disease, were not taken into account.

Click here for additional data file.

Table 1

Patient, kidney, and pancreas survival rate and human leucocyte antigens mismatch rate of patients undergoing pancreas after kidney transplant versus simultaneous pancreas and kidney transplant at 1, 3, 5, and 10 years in the included studies

First author	Total number of recipients		Patient survival rate for the following years	Kidney survival rate for the following years	Pancreas survival rate for the following years	HLA mismatch rate
	PAK	SPK
Abhinav Humar s1	205	193	1(96 versus 91):3(90 versus 90)	1(100 versus 85):3(93.6 versus 83.6):5(90 versus 76)	1(77.9 versus 78):3(61.7 versus 74.1)	1.3 versus 1.6
Ayhan Dinckan s2	34	21	5(97 versus 95.2)	5(91.2 versus 95.2)	1(79.4 versus 85.7):3(70.6 versus 71.4):5(61.8 versus 61.9)	8.8 versus 14.3
B.L.Kasiske s3	1230	4316	1(95.5 versus 96)	NA	1(82 versus 88)	NA
Bor-Uei Shyr s4	24	37	1(96 versus 97):3(96 versus 97):5(96 versus 95):10(96 versus 95)	NA	1(100 versus 100):3(100 versus 100):5(100 versus 97):10(100 versus 91)	12.5 versus 8
Edmund Huang s5	389	1402	1(100 versus 98):3(96 versus 95):5(89 versus 91)	1(100 versus 95):3(95 versus 90):5(83 versus 80)	1(81.7 versus 86):3(70 versus 81):5(53 versus 74)	8.5 versus 2.4
Farney s6	17	19	1(70 versus 84)	1(100 versus 89)	1(100 versus 95)	17 versus 19.5
Fridell JA s7	61	142	1(98 versus 95):3(90 versus 90)	1(98 versus 96):3(92 versus 90)	1(95 versus 90):3(88 versus 85)	6.6 versus 2.8
Jeffrey M. Venstrom s8	838	5379	1(95 versus 95):3(84.5 versus 87.5)	NA	NA	NA
Jens Gunther Brockmann s9	17	10	1(100 versus 100):3(100 versus 100)	1(88 versus 100)	1(88 versus 100)	NA
Jonathan A. Fridell s10	3358	12308	1(94 versus 93.7):3(88 versus 89):5(83 versus 86.4):10(63.2 versus 70)	1(93.7 versus 85.5):3(93.7 versus 85.5):5(88 versus 78):10(66 versus 61)	1(84.4 versus 87.5):3(69 versus 81.2):5(62.5 versus 75):10(44.4 versus 58.7)	14.1 versus 14.1
Muthusamy s11	15	83	NA	NA	1(80 versus 90):3(60 versus 88)	NA
M.R. Laftavi s12	23	25	1(96 versus 100):3(87 versus 88):5(87 versus 88):10(87 versus 88)	1(96 versus 96):3(87 versus 96):5(83 versus 96):10(83 versus 84)	1(96 versus 96):3(87 versus 96):5(83 versus 96):10(83 versus 84)	17.4 versus 20
M. Durlik s13	9	159	1(88.9 versus 86.2):3(88.9 versus 83):5(88.9 versus 83):10(88.9 versus 83)	1(100 versus 87):3(100 versus 85):5(100 versus 85):10(100 versus 80)	1(100 versus 71):3(100 versus 67):5(100 versus 63):10(100 versus 60)	NA
Nedo Poommipanit s14	807	5580	1(95 versus 94):3(92 versus 90):5(86 versus 86)	1(98 versus 93):3(94 versus 85):5(85 versus 77)	1(81 versus 86):3(65 versus 78):5(55 versus 73)	0.45 versus 0.08
Perosa M. s15	94	254	1(92.5 versus 81.8):3(91.5 versus 78):5(90.4 versus 75.2)	NA	1(86 versus 70):3(73.4 versus 68):5(65 versus 60)	NA
Pedro Ventura-Aguiar s16	18	139	1(100 versus 98):3(83 versus 95):5(77.8 versus 92):10(77.8 versus 90)	1(100 versus 98):3(100 versus 95):5(100 versus 90):10(100 versus 88)	1(83.3 versus 96):3(72 versus 90):5(61 versus 82):10(27.8 versus 80)	16.1 versus 2.2
Rainer WG s17	1714	6995	1(95 versus 95):3(88 versus 90)	NA	NA	NA
Robert Ollinger s18	47	442	NA	NA	1(70 versus 82):3(66 versus 80):5(62 versus 69)	NA
R.J. Stratta s19	35	162	5(86 versus 86)	5(80 versus 74)	5(66 versus 65)	7.7 versus 2.8
Sandesh Parajuli s20	24	611	NA	1(100 versus 95):3(96 versus 90):5(96 versus 85):10(96 versus 75)	31(100 versus 89):3(83 versus 85):5(83 versus 80):10(83 versus 70)	11.3 versus 0.7
Timothy S. Larson s21	47	25	1(93.6 versus 100):3(85 versus 100):5(81 versus 92)	NA	1(87 versus 92):3(81 versus 92):5(81 versus 80)	NA
Tadahiiro Uemura s22	5	17	1(100 versus 100):3(100 versus 100)	1(100 versus 100):3(100 versus 100)	1(100 versus 100):3(100 versus 100)	NA
T. Ito s23	39	232	5(97 versus 88)	NA	1(87 versus 87.5):3(64 versus 86):5(48.7 versus 82.8)	NA
Tomimaru s24	48	344	NA	NA	1(85 versus 87):3(66.7 versus 85.5):5(52 versus 83):10(42 versus 74.7)	NA

Values are n (PAK:SPK). HLA, human leucocyte antigen; PAK, pancreas after kidney transplant; SPK, simultaneous pancreas and kidney transplant; NA, not available. References to the studies can be found in Supplementary material.