Table 1.
Author, year | Study or database name | Study design | Duration of follow-up | Dialysis vintage | Dialysis type | Sample size | Allowed for non-linear relationship?* |
Was the biochemical parameter associated with mortality after multivariate adjustment? |
Graphical depiction?† | ||
---|---|---|---|---|---|---|---|---|---|---|---|
PTH | Phosphorus | Calcium | |||||||||
Studies with adequate data for meta-analysis | |||||||||||
Block 2004 [10] |
Fresenius Medical Care |
Retrospective cohort |
12-18 months (total) |
Maintenance |
HD |
40,538 |
Yes |
Yes |
Yes |
Yes |
Yes (U-shaped for calcium, linear for albumin-adjusted calcium) |
Dukkipati 2010 [29] |
DaVita-NIED |
Prospective cohort |
63 months (total) |
Maintenance |
HD |
748 |
Yes (for PTH only) |
Yes |
Not stated |
Not stated |
No |
Floege 2011 [16] |
European Fresenius Medical Care |
Retrospective cohort |
20.9 months (median) |
Incident or Maintenance |
HD |
7,970 |
Yes (also used fractional polynomials) |
Yes |
Yes |
Yes |
Yes |
Kalantar-Zadeh 2006 [14] |
DaVita |
Prospective cohort |
24 months (total) |
Maintenance |
HD |
58,058 |
Yes |
Yes |
Yes |
Yes |
Yes |
Rodriguez-Benot 2005 [24] |
Cordoba, Spain |
Prospective cohort |
97.6 months (median) |
Maintenance |
HD |
385 |
Yes (for phosphorus and PTH only) |
No |
Yes |
Yes |
No |
Tangri 2011 [30] |
UKRR |
Prospective cohort |
24 months (total) |
Incident |
HD, PD |
7,076 |
Yes |
Yes |
Yes |
Yes |
No |
Tentori 2008 [15] |
Dialysis Clinic Inc. |
Prospective cohort |
16.4 months (median) |
Maintenance |
HD |
25,588 |
Yes |
Yes |
Yes |
Yes |
Yes |
Studies that did not qualify for meta-analysis | |||||||||||
Block 1998 [31] |
USRDS (CMAS and DMMS-1) |
Retrospective cohort |
2 years |
Maintenance |
HD |
6,407 |
Yes (also analyzed as continuous linear predictor) |
No |
Yes |
Yes (text and figure are inconsistent) |
Yes (J-shaped for phosphorus, linear for calcium and PTH) |
Block 2010 [32] |
DaVita, CMS-ESRD |
Prospective cohort |
2.2 years |
Maintenance |
HD |
25,150 |
Yes |
Yes |
Yes |
Yes |
No |
Chang 2006 [33] |
VA Cooperative Study #440 |
Post-hoc RCT |
4 years |
Incident or Maintenance |
HD |
197 |
Yes |
No |
Not measured |
Not measured |
No |
Dreschler 2011 [34] |
NECOSAD |
Prospective cohort |
6 years |
Incident |
HD, PD |
1,628 |
Yes (for PTH only) |
Yes |
Not stated |
Not stated |
No |
Dreschler 2011 [35] |
NECOSAD |
Prospective cohort |
3 years |
Incident |
HD, PD |
762 |
No |
Not stated |
Not stated |
Not stated |
No |
Etter 2010[36] |
monitor! |
Prospective cohort |
562 days (median) |
Maintenance |
HD |
170 |
No |
No |
Not measured |
Not measured |
No |
Foley 1996 [37] |
Royal Victoria Hospital |
Prospective cohort |
5 years |
Incident |
HD, PD |
433 |
Yes (calcium and phosphorus were binary) |
Not measured |
No |
Yes |
No |
Ganesh 2001 [7] |
USRDS (CMAS and DMMS 1/3/4) |
Retrospective cohort |
2 years |
Maintenance |
HD |
12,833 |
Yes |
Yes |
Yes |
Not stated |
Yes (U-shaped for PTH) |
Gutierrez 2008 [38] |
ArMORR (Fresenius) |
Prospective cohort |
1 year |
Incident |
HD |
10,044 |
Yes (for phosphorus only) |
Not stated |
Yes |
Not stated |
No |
Hakemi 2010 [39] |
Tehran, Iran |
Prospective cohort |
18.4 months (mean) |
Maintenance |
PD |
282 |
Yes (for calcium and PTH only, binary) |
Yes |
Not measured |
Yes |
No |
Hsiao 2011 [40] |
Chang Gung Memorial Hospital |
Prospective cohort |
1 year |
Maintenance |
HD |
109 |
No |
No |
No |
Yes |
No |
Isakova 2009 [41] |
ArMORR (Fresenius) |
Prospective cohort |
1 year |
Incident |
HD |
10,044 |
Yes (for phosphorus only) |
Not stated |
Yes |
Not stated |
No |
Jassal 1996 [42] |
Belfast City Hospital |
Prospective cohort |
1 year |
Incident |
NR |
99 |
No |
Not measured |
Yes (but not on older subset of patients) |
Not measured |
No |
Jean 2009 [43] |
Tassin la Demi-lune, France |
Prospective cohort |
2 years |
Maintenance |
HD |
219 |
Yes (for phosphorus only, Not stated for PTH and calcium) |
Not stated |
No |
Not stated |
No |
Kalantar-Zadeh 2010 [44] |
DaVita-Los Angeles |
Prospective cohort |
5 years |
Incident or Maintenance |
HD |
139,328 |
Yes |
Yes |
Yes |
Yes |
Yes (except for PTH) |
Kimata 2007 [45] |
DOPPS 1/2 (Japan) |
Retrospective cohort |
8,056 patient-years |
Maintenance |
HD |
3,973 |
Yes (also analyzed as continuous linear predictors) |
Yes, when linear No, when categorical |
Yes, when categorical No, when linear |
Yes, for both categorical and linear |
Yes for phosphorus No for calcium or PTH |
Komaba 2008 [46] |
Takasago, Japan |
Retrospective cohort |
45 months |
Maintenance |
HD |
99 |
Yes (binary only) |
No |
Yes (but analyzed with calcium only) |
Yes (but analyzed with phosphorus only) |
No |
Lacson 2009 [47] |
Fresenius Knowledge Center |
Retrospective cohort |
1 year |
Maintenance |
HD |
78,420 |
In some models |
Yes |
Yes |
Yes |
Yes |
Leggat 1998 [48] |
USRDS (CMAS and DMMS) |
Retrospective cohort |
2 years |
Maintenance |
HD |
6,251 |
Yes |
Not measured |
Yes |
Not measured |
No |
Lin 2010 [49] |
Chang Gung Memorial Hospital |
Prospective cohort |
18 months |
Maintenance |
PD |
315 |
No |
No |
No |
No |
No |
Lowrie 1990 [50] |
National Medical Care |
Cross-sectional |
18 months |
Maintenance |
HD |
19,746 |
No |
Not measured |
Yes |
No |
No |
Lukowski 2010 [51] |
DaVita |
Retrospective cohort |
3 years |
Maintenance |
HD |
58,917 |
Yes (for phosphorus only, Not stated for PTH and calcium) |
Not stated |
Yes |
Not stated |
Yes for phosphorus No for calcium or PTH |
Maeda 2009 [52] |
Kyushu University Hospital |
Prospective cohort |
5 years |
Maintenance |
HD |
226 |
No |
No |
No |
No |
No |
McNeill 2010 [53] |
DaVita |
Cohort, unspecified |
3 years |
Maintenance |
HD |
58,917 |
Yes (for PTH only, Not stated for phosphorus and calcium) |
Yes |
Not stated |
Not stated |
Yes for PTH No for calcium or phosphorus |
Melamed 2006 [54] |
CHOICE |
Prospective cohort |
2.5 years (median) |
Incident |
HD, PD |
1,007 |
Yes |
Yes (as a time-dependent predictor) |
Yes (as a time-dependent predictor) |
Yes (as a time-dependent predictor) |
No |
Miller 2010 [22] |
DaVita-Los Angeles |
Retrospective cohort |
5 years |
Maintenance |
HD |
107,200 |
Yes |
Yes |
Yes |
Yes |
Yes for calcium No for phosphorus or PTH |
Miller 2009 [55] |
DaVita |
Cohort, unspecified |
5 years |
Maintenance |
HD |
151,555 |
Yes |
Yes |
Yes |
Yes |
No |
Morrone 2009 [56] |
MMTE (Italy) |
Prospective cohort |
18.8 months (mean) |
Incident |
HD |
411 |
Yes |
Yes |
Not stated |
Not stated |
Yes for PTH No for phosphorus or calcium |
Naves-Diaz 2011 [57] |
Fresenius-CORES |
Retrospective cohort |
4.5 years |
Incident or Maintenance |
HD |
16,173 |
Yes |
Yes |
Yes |
Yes |
Yes |
Naves-Diaz 2008 [58] |
Fresenius-CORES |
Retrospective cohort |
4.5 years |
Incident or Maintenance |
HD |
16,004 |
Yes |
Yes |
Yes |
Yes |
No |
Noordzij 2008 [26] |
NECOSAD |
Prospective cohort |
7.8 years |
Incident |
HD, PD |
1,621 |
Yes (restricted cubic splines) |
Not stated |
Yes |
Yes |
Yes (spline) |
Noordzij 2005 [23] |
NECOSAD |
Prospective cohort |
7.5 years (max) |
Incident |
HD, PD |
1,629 |
Yes |
No |
Yes |
No |
No |
Phelan 2008 [59] |
Beaumon Hospital (Ireland) |
Retrospective cohort |
4.5 years (mean) |
Incident |
HD, PD |
1,007 |
Yes (binary only) |
No |
Yes |
No |
No |
Port 2004 [60] |
DOPPS I |
Prospective cohort |
5 years |
Maintenance |
HD |
17,245 |
Yes (for phosphorus only, binary) |
Not measured |
Yes |
Not measured |
No |
Saran 2003 [61] |
DOPPS |
Retrospective cohort |
1.8-2.9 years (median, depending on region) |
Maintenance |
HD |
14,930 |
Yes (for phosphorus only, binary) |
No |
Yes |
Not measured |
No |
Shinaberger 2008 [20] |
DaVita |
Retrospective cohort |
3 years |
Maintenance |
HD |
30,075 |
Yes (restricted cubic splines) |
Not stated |
Yes |
Not stated |
Yes |
Shinaberger 2008 [21] |
DaVita |
Retrospective cohort |
3 years |
Maintenance |
HD |
34,307 |
Yes (restricted cubic splines, PTH only) |
Yes |
Not stated |
Not stated |
Yes |
Simic-Ogrizovic 2009 [62] |
Clinical Center of Serbia |
Prospective cohort |
3 years |
Maintenance |
HD |
130 |
No |
Not measured |
Yes |
No |
No |
Sit 2008 [63] |
Dicle University |
Retrospective cohort |
10 years |
Maintenance |
NR |
1,538 |
No |
Yes |
No |
No |
No |
Slinin 2005 [5] |
USRDS (1/2/3) |
Retrospective cohort |
3.9 years (mean) |
Maintenance |
HD |
14,829 |
Yes |
Yes |
Yes |
Yes |
No |
Stevens 2004 [25] |
PROMIS |
Prospective cohort |
2.6 years (median) |
Maintenance |
HD, PD |
515 |
Yes (examined interactions; also modeled continuous, linear predictors) |
No, when continuous Yes, when categorical with interactions |
Yes, for both continuous and categorical with interactions |
No, when continuous Yes, when categorical with interactions |
No |
Takemoto 2009 [64] |
Toranomon Hospital |
Prospective cohort |
8 years |
Maintenance |
HD |
68 |
No |
Yes |
No |
No |
No |
Wald 2008 [65] |
HEMO |
Post-hoc RCT |
6.5 years |
Maintenance |
HD |
1,846 |
Yes |
No |
Yes |
Yes |
Yes for calcium and phosphorus No for PTH |
Young 2004 [66] | DOPPS 1/2 | Retrospective cohort | 2 years | Maintenance | HD | 15,475 | No | Yes | Yes | Yes | No |
HD, hemodialysis; NR, not reported; PD, peritoneal dialysis; PTH, parathyroid hormone; RCT, randomized controlled trial.
* Study explicitly allowed for a non-linear relationship between biochemical parameters and mortality (eg, use of a categorical variable or splines).
†Study provided a graphical depiction of the relationship between biochemical parameters and mortality.