Alicja E. Grzegorzewska, Magdalena Leander
From: Chair and Department of Nephrology, Karol Marcinkowski University of Medical
Sciences, Poznan, Poland.
The present study evaluated whether estimation of lymphocyte
subset counts can be more helpful than total lymphocyte count (TLC) in earlier
diagnosis of immune and nutritional changes in the course of continuous ambulatory
peritoneal dialysis (CAPD). For the study, 50 CAPD patients were divided into
four groups depending on dialysis duration. Group I consisted of patients
treated for 6 – 12 months (n = 15); group II, for 13 –
24 months (n = 16); group III, for 25 – 36 months (n =
12); and group IV, for more than 36 months (n = 7). Thirteen patients,
being 8 ± 7 days before CAPD initiation, were included in group 0.
Flow cytometry was used for estimation of lymphocyte subsets (determination
of CD3, CD4, CD8, CD19, and CD16+56 antigens).
Our uremic patients started CAPD therapy with decreased TLC and lymphocyte subset
(excluding CD16+56) counts. After 6 – 12 months of CAPD therapy, a
significant increase in TLC, CD4:CD8 ratio, and all examined lymphocyte subset
counts was observed. In the next years of CAPD therapy, TLC, CD3, CD4, CD8,
and CD19 cell counts decreased. In patients on CAPD for more than 36 months,
CD3, CD4, CD8, and CD19 cell counts were below the normal range, but mean TLC
was maintained in the normal range, and CD16+56 exceeded the upper limit of
normal. A significant negative correlation between CD19 cell count and dialysis
duration was seen (r = –0.298, p = 0.035, n = 50).
In conclusion, the first months of CAPD therapy see an improvement in immune
and nutritional status as expressed by an increase in TLC, lymphocyte subset
counts, and CD4:CD8 ratio. Repeat determinations of CD3, CD4, CD8, and CD19
cell counts indicate that these counts decrease earlier than an evaluation of
TLC indicates.
We recommend lymphocyte subset determinations for detection of immune and nutritional
abnormalities in the course of CAPD treatment. An increase in natural killer
cells above the normal range may reflect chronic sterile or infectious inflammatory
response, which deteriorates nutritional status.
A decrease in total lymphocyte count (TLC) during the course of continuous
ambulatory peritoneal dialysis (CAPD) is an indicator of disturbances in immune
response and nutritional status (1–3) and is mentioned as a prognostic
index of mortality in CAPD patients (4–6).
In our studies, a significant decrease in TLC was seen at months 36 –
39 of CAPD treatment as compared with values obtained during the first 3 months
of CAPD. Still, this mean decreased TLC was above the lower normal limit (2,7).
Mean TLC below the normal range was observed in our patients at 52 –
54 months of CAPD therapy (Grzegorzewska AE, Leander M. Unpublished data). Studies
by Palop et al (8) revealed a significant decrease not only in TLC, but also
in B lymphocyte count during the course of CAPD as compared with the start
of treatment. Our preliminary data indicate that lymphocyte subset counts (except
natural killer cells) decrease over the course of CAPD treatment, and that this
decrease can be seen earlier than the fall in TLC (9).
In the present study, we undertook an extended evaluation to determine whether
estimation of lymphocyte subset counts can be more helpful than TLC in earlier
diagnosis of immune and nutritional changes during the course of CAPD.
The study was carried out in 63 adult uremic patients. The underlying kidney
diseases were chronic glomerulonephritis (21 cases), diabetic nephropathy
(14 cases), chronic pyelonephritis (12 cases), hypertensive nephropathy
(9 cases), polycystic kidney disease (4 cases), gout nephropathy (1 case),
and unknown etiology (2 cases).
Of the 63 patients, 50 already being treated with CAPD were divided into 4 groups,
depending on dialysis duration. Group I consisted of patients treated for
6 – 12 months (9.3 ± 1.6 months; n = 15: 3 women,
12 men; age: 54.5 ± 13.4 years); group II, for 13 –
24 months (18.0 ± 2.7 months; n = 16: 5 women, 11 men; age:
49.9 ± 14.3 years); group III, for 25 – 36 months
(28.9 ± 3.2 months; n = 12: 6 women, 6 men; age: 49.6 ± 8.3
years); and group IV, for more than 36 months (42.2 ± 5.3 months;
n = 7: 2 women, 5 men; age: 52.7 ± 10.3 years). Another 13
patients (6 women, 7 men; age: 55.4 ± 15.5 years), being 1 –
22 days (8 ± 7 days) before CAPD initiation, were included
as group 0. All group 0 patients had a permanent peritoneal catheter
implanted and were waiting for the start of CAPD.
Patients in group 0 were not receiving erythropoietin. In groups I
to IV, the percentages of patients who were taking recombinant human erythropoietin
subcutaneously concomitantly with an oral drug containing iron and folic acid
were 40%, 56%, 67%, and 71%, respectively. The erythropoietin doses, calculated
for the entire CAPD course in patients receiving erythropoietin, were 2884 U,
3177 U, 8590 U, and 4915 U per patient–month, respectively.
Other drugs and supplements were administered depending on clinical indications.
Percentages of lymphocyte subsets were determined by flow cytometry, using the
commercially available monoclonal antibodies CD3, CD4, CD8, CD19, CD16+56 (Becton-Dickinson,
San Jose, CA, U.S.A.). Granulocytes and monocytes were excluded from the calculations
with the help of monoclonal antibodies CD45 and CD14 (Becton-Dickinson). Absolute
counts of lymphocytes were calculated using white blood cell count, estimated
by routine procedure (stained smears, chamber method). In our laboratory, the
normal range for TLC is 1.5 – 3.5×109/L; for CD3 cells (pan
T cells), 1.1 – 1.7×109/L; for CD4 cells (helper T cells),
0.7 – 1.1×109/L; for CD8 cells (cytotoxic-suppressor T cells),
0.5 – 0.9×109/L; for CD19 (B cells), 0.2 – 0.4×109/L;
and for CD16+56 cells [natural killer (NK) cells], 0.2 – 0.4×109/L.
The normal CD4:CD8 ratio is 0.8 – 2.2.
Means and standard deviations for TLC and for each of the examined cell subsets
were calculated for all patient groups and were compared to their respective
normal ranges and to each other using the Mann–Whitney test. Correlations
between dialysis duration and lymphocyte counts were checked using the Spearman
test. All results of groups 0 to IV, as well as those of groups I to IV,
were used for correlation.
Table I presents the results of the estimations of TLC, lymphocyte subsets,
and CD4:CD8 ratio. The TLC and all examined lymphocyte subsets (except CD16+56
cells) were below the normal range immediately before CAPD initiation. After
initiation of CAPD, a significant increase in TLC and lymphocyte subset counts
was seen. After 6 – 12 months of CAPD therapy, TLC and all lymphocyte
subsets were in the normal range. In the subsequent years of CAPD therapy, TLC
and CD3, CD4, CD8, and CD19 cell counts decreased (usually insignificantly).
In patients on CAPD for more than 36 months, CD3, CD4, CD8, and CD19 cell counts
fell below the normal range, but mean TLC remained in the normal range, and
CD16+56 exceeded the upper normal limit. A significant negative correlation
was seen between CD19 cell count and dialysis duration (r = –0.298,
p = 0.035, n = 50). Mean values of the CD4:CD8 ratio were in the normal
range in all studied groups; but, in groups I and II, the ratio was significantly
higher than in group 0.
No significant correlation was seen between lymphocyte counts and dialysis duration
when the results of all groups were analyzed. After excluding the results obtained
in group 0, a significant negative correlation was found between CD19 count
and dialysis duration (r = –0.298, p = 0.035, n = 50).
The examined patients started CAPD therapy with decreased TLC and lymphocyte
subset counts (except CD16+56 cells). The TLC, all cell subpopulations, and
the CD4:CD8 ratio increased during the first year of CAPD therapy.
In patients grouped according to time on hemodialysis, it has been shown that
subnormal TLC at the beginning of dialysis becomes normal over time on treatment
(10). Improvement in lymphocyte counts might be due both to the anabolic effects
of the first few months of dialytic therapy (11,12) and initiation of treatment
with recombinant human erythropoietin. Most data (13–15) indicate a beneficial
effect of erythropoietin on TLC or lymphocyte subset count. According to Gelfand
et al (16) the CD4:CD8 ratio increases even up to 24 months of CAPD treatment.
Despite the usual practice of increasing the dose of erythropoietin during the
course of CAPD therapy, a worsening of immune and nutritional parameters is
observed with prolongation of dialysis treatment, including a decrease in lymphocyte
count (1–8) and in CD4:CD8 ratio (16). In patients on maintenance hemodialysis,
increased lymphocyte apoptosis has been shown, leading to a decreased number
of lymphocytes in the peripheral blood (17,18). Altered numbers of immune cells
contribute to immunologic abnormalities (3), depressed erythropoiesis (19),
increased infection rate (20), and poor outcome (4). Early detection of immunologic
disturbances and features of malnutrition may initiate clinical intervention,
resulting in more effective treatment with peritoneal dialysis. Repeated determination
of B and T lymphocyte subset counts seems to be helpful in the early diagnosis
of such disturbances.
The study by Palop et al (8) indicates that the B lymphocyte count decreases
during the course of CAPD sooner than does the TLC. Our results demonstrate
that numbers of B lymphocytes may decrease from the beginning of CAPD treatment.
In this situation, CAPD treatment cannot normalize the B lymphocyte count.
Earlier or even “healthy” start of CAPD therapy may possibly abolish
the detrimental effect of pre-dialysis undernutrition.
The T lymphocyte subpopulation composition (CD3, CD4, CD8) is sensitive
to influences of sex, age, hormones, protein or trace element intake, physical
effort, stress, and circadian rhythm (21). Administration of angiotensin converting
enzyme inhibitors may decrease T lymphocyte proliferation, causing an increase
in serum bradykinin concentration (22). In our study, lymphocytes bearing antigen
CD8 or CD4 were below the normal range in all examined patient groups; but,
after initial normalization related to the beginning of CAPD therapy, the CD3
cell count fell below the normal range after 3 years of CAPD treatment. The
decrease was not accompanied by a fall in TLC. In patients on maintenance hemodialysis
for 3.6 ± 2.1 years, a significant decline in TLC, in absolute
numbers for all T lymphocyte subsets, and in CD4:CD8 ratio has been observed
(23).
Natural killer cells are derived from bone marrow. They play a role in defense
against infection (24). Their cytotoxic function is bone marrow and cytokine
dependent (25,26). In the examined patients, we observed an increase in the
number of natural killer cells over the normal range with length of time on
CAPD. The increase was distributed according to the duration of CAPD. This finding
accords with the study performed in CAPD patients by Palop et al (3). Natural
killer cells and their precursors are more resistant to immunosuppressive agents
than are other immunocompetent cells (27,28), so that suppressive uremic toxins
affect the T lymphocyte number and function more effectively than they
affect the number and function of natural killer cells (28,29). In patients
on peritoneal dialysis, episodes of peritonitis, exit-site infection, and tunnel
infection may stimulate production of natural killer cells. In hemodialyzed
patients, the percentage of natural killer cells was significantly higher as
compared with a control group, but the absolute number of these cells was analogous
to those found in healthy volunteers (23,30). This difference between peritoneal
and hemodialysis patients might be related to the greater number of infectious
episodes in peritoneal dialysis patients.
A significant correlation between CAPD duration and TLC or lymphocyte subset
counts was shown only for CD19 cells, when results obtaining during CAPD treatment
were taken for analysis (groups I to IV). An increase in lymphocyte counts
after initiation of CAPD treatment disturbed the correlation results. Our study
indicates that CD19 cells are especially sensitive to the abnormalities related
to prolonged CAPD therapy.
| table i Total lymphocyte count (TLC) and lymphocyte subset
counts in peripheral blood from uremic patients grouped according to duration
of continuous ambulatory peritoneal dialysis (CAPD). |
|||||
| |
Immediately before CAPD initiation |
CAPD duration (months)
|
|||
| 6–12 | 13–24 | 25–36 | 37–52 | ||
| Lymphocytes (109/L)
|
(n=13)
|
(n=15)
|
(n=16)
|
(n=12)
|
(n=7)
|
|
TLC |
1.16±0.71a | 1.91±0.72b | 1.56±0.66 | 1.58±0.50 | 1.81±0.85 |
|
CD3 |
0.86±0.53a | 1.37±0.57b | 1.15±0.52 | 1.15±0.38 | 1.07±0.53a |
|
CD4 |
0.33±0.23a | 0.62±0.31a,b | 0.53±0.23a | 0.50±0.21a | 0.41±0.21a |
|
CD8 |
0.31±0.21a | 0.37±0.20a,b | 0.34±0.22a | 0.35±0.16a | 0.34±0.17a |
|
CD19 |
0.08±0.09a | 0.16±0.10a,b | 0.09±0.04a,c | 0.11±0.08a | 0.10±0.07a |
|
CD16+56 |
0.20±0.16 | 0.35±0.15b | 0.30±0.16 | 0.28±0.12 | 0.44±0.41d |
|
CD4:CD8 ratio |
1.14±0.42 | 1.81±0.74b | 1.81±0.83b | 1.63±0.92 | 1.31±0.43 |
|
a Mean below normal range. |
|||||
The first months of CAPD therapy produce an improvement in immune and nutritional
status as expressed by an increase in TLC, lymphocyte subset counts, and CD4:CD8
ratio. Repeated determinations of CD3, CD4, CD8, and CD19 cell counts indicate
that these counts decrease earlier than does TLC.
We recommend lymphocyte subset determinations for detection of immune and nutritional
abnormalities in the course of CAPD treatment. An increase in natural killer
cells above the normal range may reflect chronic sterile or infectious inflammatory
response, which deteriorates nutritional status.
Alicja E. Grzegorzewska, md phd, Chair, Department of Nephrology, Al. Przybyszewskiego
49, 60–355 Poznan, Poland.