Quality of Life in CAPD, Transplant, and Chronic Renal Failure Patients with Diabetes

Abstract

Although quality of life (QoL) is an important outcome measure, there are few studies of QoL in diabetic patients. We performed a cross-sectional study to assess QoL in such patients comparable for age, sex, and co-morbidity. Patients. Group CAPD: DM (n = 19, 12 males), diabetic CAPD patients; group CAPD: no DM (n = 26, 15 males) CAPD patients without diabetes; group TXP (n = 20, 10 males), diabetic transplant patients; and group CKD: DM (n = 20, 11 males), diabetics with chronic kidney disease. Two valid QoL instruments were used: a generic one (SF-36) and one that is disease-specific (RQLP). Results. As a whole, CAPD patients scored badly as far as concerned the physical domain, but with a good mental adaptation. Diabetic CAPD patients exhibited worse QoL for physical functioning, energy, vitality, leisure activity, and eating/drinking limitations. Diabetic transplant patients exhibited the best QoL. The RQLP instrument had better discriminative power for domains such as eating/drinking, treatment effects, and psychosocial aspects. Using analysis of co-variance and adjusting for age, sex, and co-morbidity, QoL differences disappeared.

In conclusion, diabetic CAPD patients exhibited the worst QoL though with a satisfactory mental adaptation, a renal-specific instrument had better discriminative power, and the prevention of co-morbidity is likely to improve QoL in such patients.

Keywords:

• quality of life
• diabetes mellitus
• CAPD
• transplantation
• chronic kidney disease

INTRODUCTION

In the last two decades, quality of life has become an increasingly important outcome measure in medicine.[1–3] Over this period, the treatment of patients in end-stage renal disease (ESRD) has also changed dramatically with the development of newer but expensive renal replacement techniques and a greater acceptance of higher risk patients, such as those with diabetes. Diabetic nephropathy now accounts for more than 35% of all patients on dialysis in Europe[4] and 50% in the United States.[5] It is known that diabetes is accompanied by multi-system and organ dysfunction, and serious questions have been raised about the outcomes (survival, morbidity) in this group of patients, in particular the impact of treatment on their quality of life.[1],[6],[7]

There are many studies of quality of life in ESRD patients, but a limited number in patients with diabetes.[1] In the absence of prospective and longitudinal studies, quality of life in diabetic CAPD patients has only partly been examined by assessing co-morbid factors and survival, using mainly generic instruments.[1],[8–11] In addition, there is disagreement about the methodology and which instrument to use.[1],[12] Many instruments for assessing quality of life are now available.[1],[10] CAPD is a well-established mode of renal replacement therapy and for many the treatment of choice for those with diabetes, so the need for monitoring and follow-up of quality of life in these patients is of great importance.

The aims of this study are as follows:

• to assess the quality of life of patients with diabetes undergoing CAPD treatment using two validated instruments, the generic Health Survey SF-36 [13–15] and the renal-specific Renal Quality of Life Profile (RQLP),[16–18]

• to compare the results of these findings with other comparable patient groups, such as non-diabetic CAPD patients, patients with diabetic chronic renal failure (CRF) not on dialysis, and diabetic renal transplant patients,

• to assess the feasibility of using these quality of life instruments for routine care (i.e., assessing their QoL periodically), and especially

• to assess patient's perception to what really matters for his/her QoL.

Information from this study will be used to modify future policy and treatment modalities in an attempt to achieve better outcomes and improve the quality of life for this high-risk group of patients.

SUBJECTS AND METHODS

Study Population

The study population consisted of 176 adult patients who were selected by a computer-based program to be comparable for age, sex, type, duration of diabetes (if present), duration of dialysis (if any), marital status, and co-morbidity. They were distributed as follows:

• CAPD patients with diabetes (CAPD: DM, 39 patients, 21 males). These patients represented the total number of diabetic CAPD patients (due to diabetic nephropathy, 20% of all CAPD patients), followed by the CAPD Clinic of the Renal Unit of Manchester Royal Infirmary (MRI).

• CAPD patients without diabetes (CAPD: no DM, 50 patients, 26 males).

• Transplant patients with diabetes (TXP, 38 patients, 21 males). These are renal patients suffering from diabetic nephropathy, excluding those with secondary diabetes, and a mean serum creatinine of 186 ± 44 mmol/L (range 104–320 mmol/L). This group represented the whole number of transplanted patients with diabetes followed by the joined Diabetes and Renal Clinic of MRI.

• Patients with diabetic nephropathy and chronic kidney disease (CKD: DM, 50 patients, 30 males). These are patients with a mean serum creatinine 209 ± 88 mmol/L, range 130–500 mmol/L.

Inclusion criteria were as follows: patients were aged over 16 years, those on CAPD had to be on this treatment for at least three months and in a stable clinical condition for at least one month before completing the questionnaire. Exclusion criteria were as follows: patients aged < 16 years, the presence of acute psychiatric or medical illness within one month prior to entry in the study, and patients unable to understand and complete the questionnaire.

The Central Manchester Ethics Committee approved the study, and all patients gave written informed consent prior to taking part in the study. All patients were mailed, or given to self-complete, two instruments of measuring quality of life items, each measuring the impact of certain quality of life determinants in the patients' daily living. The instruments were:

• the Renal Quality of Life Profile (RQLP), a renal disease-specific instrument based on constructs representing renal patients' own quality of life determinants.[16–18] It comprises 43 items describing health-related dysfunctional behaviors that are specific to ESRD patients, grouped into five categories (i.e., eating/drinking, RQLP A; physical activity, RQLP B; leisure, RQLP C; psychosocial aspects, RQLP D; and treatment effects, RQLP E). A five-point Likert scale score is used where an increase in impairment is seen as an increase in overall score. Scale scores were transformed to a 0–100 scale. The instrument is comprehensive, brief, simple to administer, easy to score, and has shown reliability and content and face validity with sensitivity to changes of health.[16–18]

• The second questionnaire is an objective instrument, the 36-item MOS Short Form Health Survey Questionnaire (SF-36).[13–15] The SF-36 is a generic multidimensional instrument consisting of eight multi-item scales representing physical functioning, role limitations due to physical problems, role limitations due to emotional problems, social functioning, mental health, energy/vitality, bodily pain, and general health perceptions. Patients are also asked to rate the effect of their health on their life during the past four weeks to assess the possible change of health during that period. Scale scores were transformed to a 0–100 scale, with scores of each domain from 1–5, where an increase of the score is seen as a more favorable quality of life. The SF-36 has been reliable and valid in various demographic and patient populations, including ESRD patients.[11],[19],[20]

All patients were approached in the outpatient clinics and received an information sheet explaining the details of the study. The same medical team assessed their clinical status. The possibility of accompanying neuropathy of patients was assessed, and measurements were done of the Neuropathy Disability Score (NDS), the Vibration Perception Threshold (VPT), and the Neuropathic Symptom Score (NSS).

The Neuropathy Disability Score was based on an examination of the big toe for pain using a pinprick, vibration feeling using a 128 Hz tuning fork, temperature of the dorsum of the foot, and Achilles reflexes. Scores for each examined domain were from 0 (normal) to 1 (abnormal), with a maximum score of 10 (5 in each leg) and minimum of 0.[21],[22] Vibration perception threshold was measured three times with a neuro-biothesiometer, recording the mean final value in Volts, and comparing it with a scale of normal age matched values.[21],[22] The Neuropathic Symptom Score is based upon the patient's description of the feelings of fatigue, cramping, burning, numbness, and tingling, with maximum scores where the feet were affected. In addition, scores depend on the time when the symptoms are most severe (nighttime symptoms score highest), wakening from sleep, and relief of pain by walking (characteristic of diabetic neuropathy).[21],[22]

Erythropoietin treatment, dialysate composition, frequency of peritonitis, hospitalization other than for peritonitis, serum intact parathyroid hormone (iPTH), hemoglobin, and serum bicarbonate were also evaluated. Co-morbidity was defined in terms of the presence of conditions related to diabetes (when present) and atherosclerosis. Such co-morbid factors included retinopathy, blindness, diabetic neuropathy, ischemic heart disease (history of angina), previous myocardial infarction, neuropathic foot ulcer, cerebrovascular accident, peripheral vascular disease, and history of amputation.

CAPD patients were dialyzed with four two-liter exchanges per day, with a dialysate calcium composition of 1.25 mmol/L.

Statistical Procedures

Data were analyzed by one-way analysis of variance for differences of variables between groups. The Spearman correlation coefficient was used to correlate domains between the instruments and the Kruskal-Wallis test for statistical differences between each subscale of the domains of the quality of life instruments. Analysis of co-variance for the influence of confounding variables was performed, and adjustments were made. Numeric data are presented as means ± SD. Statistical significance was defined as p < 0.05.

RESULTS

Demographic and Laboratory Data

Out of the 176 patients that were selected for the study, a total of 85 (response rate 48%) completed and returned the QoL questionnaires. They were distributed as follows, as seen in Table 1: nineteen (12 males) in Group CAPD: DM, 26 (15 males) in Group CAPD: no DM, 20 (10 males) in Group TXP, and 20 in Group CKD: DM (11 males). Patients who did not return the questionnaires were comparable with those that answered for clinical status, co-morbidity, duration and control of diabetes (if present), sex, race, and age. The main reason they did not want to complete the questionnaires was that they were involved in other studies. Some claimed personal reasons as well.

Table 1 Demographic data of all study groups

	CAPD: DM	CAPD: no DM	TXP	CKD: DM	p
Number of patients	19	26	20	20	NS
Age (years, M ± SD)	54.2 ± 13.9	53.6 ± 12.7	49.2 ± 11.4	53.4 ± 8.6	NS
Sex (M/F)	12/7	15/11	10/10	11/9	NS
Marital status (M/S/D/W), %	78/11/6/5	65/16/8/11	89/11/−/−	92/8/−/−	NS
Employment (E/UE/HW/R), %	22/33/6/39	46/11/19/24	50/−/17/33	42/17/25/16	NS
Type of diabetes (I/II), n	13/6	NA	15/5	7/13	0.03
Duration of diabetes (months)	21.5 ± 7.5	NA	23.1 ± 9.4	22.7 ± 8.9	NS
Treatment of diabetes (I/O/D), n	17/1/1	NA	19/−/1	14/6/−	NS
Duration of dialysis (months)	12 ± 10	24 ± 19	NA	NA	0.01
Hb (g/dL)	10.1 ± 1.2	11.1 ± 1.4	12.7 ± 1.8	11.5 ± 1.7	< 0.001
HbA1c (%)	8.3 ± 1.3	NA	8.7 ± 1.4	9 ± 1.3	NS
Urea (mmol/L)	22 ± 7.5	21 ± 4.5	13 ± 5.5	16 ± 6	< 0.001
Creatinine (mmol/L)	690 ± 188	856 ± 231	171 ± 45	209 ± 89	< 0.001
Albumin (g/L)	35 ± 3.4	37 ± 5.4	41 ± 2.9	41 ± 4.8	< 0.001
Ca (mmol/L)	2.5 ± 0.1	2.6 ± 0.1	2.4 ± 0.1	2.3 ± 0.1	< 0.001
PO4 (mmol/L)	1.6 ± 0.6	1.7 ± 0.4	1 ± 0.2	1 ± 0.3	< 0.001
HCO3 (mEq/L)	26 ± 2.6	27 ± 3.4	No data	27 ± 2.9	NS
Proteinuria (g/day)	NA	NA	No data	2.6 ± 2.6	NA
EPO treatment (Y/N)	8/11	11/15	NA	NA	NS
VPT	30 ± 10	13 ± 4	21 ± 11	28 ± 11	< 0.001
NDS	7.3 ± 2.8	1.8 ± 1.5	3.6 ± 2.6	6.7 ± 4.1	< 0.001
NSS	6.3 ± 2.1	1.6 ± 4	3.6 ± 3.8	4.3 ± 4.5	< 0.001
History of transplant (Y/N)	0/19	7/26	N	N	0.01

Abbreviations: M = married, S = single, D = divorced, W = widow/er, E = employed, UE = unemployed, HW = housekeeping, R = retired, Y = Yes, N = No, NA = Non-applicable, I = insulin, O = oral agents, D = Diet; for VPT, NDS, NSS see text.

There were no significant differences between the groups of responders in terms of age, sex, employment and marital status, duration, and control of diabetes (if present). On the other hand, responders of Group CKD: DM contained more patients with type II diabetes (p < 0.05), while responders of Group CAPD: no DM had been on dialysis for a longer period of time than Group CAPD: DM (p < 0.05). There were no significant differences between diabetic patients in terms of the presence of retinopathy, blindness, neuropathy, history of foot ulcer, cerebrovascular accident, peripheral vascular disease, or the presence of amputation (see Table 2). There was a tendency for CAPD patients with diabetes to have a higher incidence of ischemic heart disease, previous myocardial infarction, and peripheral vascular disease (p ≥ 0.07, 0.09, and 0.08, respectively), but with no statistical significance. Of note, the control group without diabetes (CAPD: no DM) had a similar frequency of ischemic heart disease, myocardial infarction, and peripheral vascular disease due to atherosclerosis. Neuropathy was more severe in diabetic CAPD patients (p < 0.05). The values of glycated hemoglobin (HbA1C) were not statistically different between groups, suggesting similar glycemic control through time (see Table 1). Serum albumin levels were lower in CAPD patients (groups CAPD: DM and CAPD: no DM), with a worse value in CAPD: DM group (p < 0.001). Hemoglobin levels were low in the CAPD patients, with a higher value in TXP group (p < 0.001, see Table 1). There was no difference in the use and dose of erythropoietin between the CAPD groups. Group CAPD: DM had lower serum creatinine levels than group CAPD: no DM (p ≥ 0.03), while groups TXP and CKD: DM had similar values (see Table 1). Urea levels between CAPD: DM and CAPD: no DM patients were similar, while there was a tendency in CKD: DM group toward higher levels of urea in comparison with group TXP. Serum calcium and phosphate levels were higher in diabetic and non-diabetic CAPD patients (p < 0.001).

Table 2 Presence of co-morbidity in all study groups

	CAPD: DM	CAPD: no DM	TXP	CKD-DM	p
Number of patients	19	26	20	20	NS
Diabetic retinopathy (Y/N)	17/2	NA	17/3	17/3	NS
Blindness (Y/N)	4/15	NA	1/19	2/18	NS
Diabetic neuropathy (Y/N)	15/4	NA	13/7	16/4	NS
Ischaemic heart disease (Y/N)	10/9	5/21	4/16	6/14	0.07
History of MI (Y/N)	3/16	4/22	0/20	1/19	NS
History of foot ulcer (Y/N)	4/15	NA	4/16	4/16	NS
Cerebrovascular accident (Y/N)	3/16	4/22	2/18	1/19	NS
Peripheral vascular disease (Y/N)	6/13	3/23	1/19	2/18	0.09
History of amputation (Y/N)	1/18	NO	1/19	NO	NS

Abbreviations: Y/N = Yes/No, NA = non-applicable.

Quality of Life Survey

Seven patients in the CAPD: no DM group, though none in CAPD: DM group, had previously undergone renal transplantation. This suggests that the selection existed between these groups in that patients without diabetes were more likely to have undergone renal transplantation than CAPD patients with diabetes, despite the fact that other aspects of co-morbidity were similar between the groups (see Table 2). Nearly 50% of CAPD patients had stopped working because of ill health, while only 30% of transplant patients and 10% of chronic kidney disease patients with diabetes had done so. Only 16% of CAPD: DM patients, and 19% of CAPD: no DM patients, believed they led a normal life, compared to 61% in TXP group and 53% in CKD: DM group (see the RQLP instrument in Table 3).

Table 3 Scoring results of all groups: General questionnaire SF-36

SF-36	CAPD: DM	CAPD: no DM	TXP	CKD: DM	p
Physical functioning	23.7 ± 5	36 ± 4.7	51.5 ± 6.6	46.4 ± 8.4	0.01
Role limitation (physical problems)	16.2 ± 8	12 ± 3.8	47.5 ± 10	47.5 ± 0.2	0.001
Role limitation (mental)	33.3 ± 11	34.7 ± 8.6	76.7 ± 9	55 ± 10.6	0.007
Social functioning	42.1 ± 8.2	49.3 ± 4.7	76.7 ± 5	66.7 ± 6.8	0.001
Mental health	61 ± 5.8	61 ± 3.5	70.6 ± 4.4	63.8 ± 5.8	0.45
Energy and vitality	29.5 ± 5.1	32.4 ± 3.7	52.8 ± 4.4	45.8 ± 6.6	0.003
Pain	52.6 ± 7.8	52.4 ± 5.9	64.4 ± 5.3	49.4 ± 7.1	0.40
Health perceptions	23.3 ± 5.6	35 ± 4.5	39.2 ± 5	34 ± 6	0.20
Change in health	52.8 ± 5	52.9 ± 4.5	50 ± 4.3	52.5 ± 5.1	0.97

Note. Highest score = better quality of life.

^* The p value is for a one-way ANOVA between the four groups.

SF-36 Instrument

All four groups had similar perceptions of their health (p ≥ 0.20). In all groups of patients, there was no significant reported change in their health during the four weeks prior to completion of the questionnaires, suggesting stability of their condition at that time. For physical function, role limitation due to emotional problems, social functioning, and energy/vitality domains, scores were worst in group CAPD: DM and best for group TXP (see Table 3). Role limitation due to physical problems was more severe than that due to emotional problems, and again was most marked in CAPD patients (see Table 3). The presence of bodily pain was similar in all groups (p = NS) despite their clinical and quality of life differences, suggesting that this domain was not affecting their attitude to life.

Mental Health

No differences were found between groups with regard to attitudes toward their health problems, and indeed all groups were generally positive in outlook suggesting good mental adaptation.

Renal Quality of Life Profile (RQLP)

The data from the RQLP can be seen in Table 4.

Table 4 Renal specific questionnaire (RQOLP, scoring of all groups)

RQOLP	CAPD: DM	CAPD: no DM	TXP	CKD: DM	p
Eating/drinking	51.5 ± 26	47.5 ± 21.7	13.8 ± 14	29.4 ± 24	0.001
Physical activity	71 ± 26	57 ± 19	35 ± 27	53.4 ± 36	0.003
Leisure activity	80 ± 25	63.3 ± 24	20.2 ± 24	34 ± 36	0.001
Psychosocial aspects	50 ± 31	45 ± 19	26 ± 16	39 ± 33	0.01
Treatment effects	45.5 ± 32.2	47.2 ± 18.2	33.7 ± 22.9	40.4 ± 32	0.31
Normal life (Y/N)	3/16	5/21	11/7	10/9	0.003
Stop working (Y/N/NA)	9/6/4	12/10/4	6/11/3	2/10/6	0.03

Higher score = worse quality of life.

Although each subscale of RQOLP questionnaire was tested (with Kolmogorov-Smirnov test) and found normally distributed, because of the high variation of the standard deviation, the comparisons between groups were made using the Kruskal-Wallis test.

Eating/Drinking Domain

Drinking limitations were worst for CAPD patients, as might be expected, but CAPD: DM patients reported even more limitations in this domain than the CAPD: no DM group. However they appeared otherwise satisfied with their diet. The TXP and CKD: DM groups reported fewer limitations, with transplant patients having the best score once again. TXP patients also had the best scores for physical and leisure activities, with the worst scores seen for CAPD: DM (p ≥ 0.004 and p ≥ 0.001, respectively).

Psychosocial Domains

Daily activities and difficulties in social life were the domains scored lowest by CAPD patients, especially CAPD: DM patients (p ≥ 0.001). However, they did not score poorly in facing financial matters, thoughts for the future, problems with sexual and family life, or self-esteem. Renal transplant patients again had the best scores in this domain.

Treatment Effects

Overall, there were no differences in the way that groups perceived the effects of their treatment on quality of life (p = NS). However, CAPD patients reported much preoccupation with the existence of pruritus and their skin condition (p ≥ 0.003 and 0.01, respectively).

Correlation Coefficients between the Two Questionnaires

There was a good correlation between subscales of the two questionnaires when they were measuring similar items (see Table 5), such as physical activity (r = 0.73). Between the two questionnaires, social functioning and social role correlated well with psychosocial aspects and leisure and physical activity (r = 0.63 and 0.68 respectively); energy and vitality with physical activity (r = 0.60); and mental health with psychosocial aspects (r = 0.64); indicating that both were measuring what was intended to measure. Other items correlated very poorly, and some domains, generally accepted as important to renal patients (eating/drinking habits, effects of treatment on their quality of life, skin condition, pruritus), were not measured by the generic SF-36 instrument (see Table 5). Data of these correlations also provide additional evidence of validity of RQLP instrument, given that the SF-36 is a well-known valid instrument that is frequently used. Correction for co-morbidity made the differences between groups disappear (see Table 6).

Table 5 Spearman correlation coefficients in the whole sample of patients between the study domains of the two instruments

	RQOLP
SF-36	Eating/drinking	Physical activity	Leisure	Psychosocial aspects	Treatment effects
Physical functioning	0.2	0.73	0.52	0.46	0.38
Role limitation (physical)	0.27	0.5	0.46	0.39	0.3
Role limitation (mental)	0.2	0.37	0.36	0.4	0.24
Social functioning	0.4	0.6	0.64	0.68	0.55
Mental health	0.28	0.52	0.41	0.64	0.57
Energy/vitality	0.33	0.6	0.58	0.52	0.42
Pain	0.2	0.66	0.38	0.5	0.53
Health perceptions	0.29	0.64	0.43	0.58	0.58
Change in health	0.22	0.06	0.03	0.1	0.03

Table 6 Results of analysis of co-variance for comparisons of all patients, with the use of both instruments in respect to subscales of all their domains, adjusted for co-morbidity, age, sex, duration of dialysis, neuropathy, and laboratory variables

Variable	No adjustment	Adjustment for age, sex, co-morbidity, dialysis duration	Adjustment for age, sex, co-morbidity, dialysisduration, and VPT	Adjustment for age, sex, co-morbidity, dialysis duration, VPT, and laboratory variables
RQLP A	<0.0005	<0.0005	0.0005	0.74
RQLP B	0.001	0.069	0.12	0.77
RQLP C	<0.0005	<0.0005	<0.0005	0.083
RQLP D	0.024	0.18	0.17	0.83
RQLP E	0.35	NA	NA	NA
SF-36 physical activity	0.017	0.091	0.057	0.50
SF-36 role limitation (physical)	0.001	0.002	<0.0005	0.48
SF-36 role limitation (mental)	0.007	0.026	0.069	0.89
SF-36 social functiononig	0.001	0.005	0.001	0.89
SF-36 mental health	0.45	NA	NA	NA
SF-36 energy/vitality	0.003	0.020	0.001	0.18
SF-36 bodily pain	0.40	NA	NA	NA
SF-36 health perceptions	0.20	NA	NA	NA
SF-36 health change	0.97	NA	NA	NA

Abbreviation: NA = Non-applicable because p value (adjusted) was NS. From the above table, it is seen that adjustment for age, sex, comorbidity, duration of dialysis, and neuropathy (VPT) showed differences between groups for RQLP B and D, and physiacl activity and role limitation (mental) disappear. Moreover, additional adjustment for laboratory variables made all statistical differences disappear.

RQLP A,B,C,D,E are refering to RQLP domains (see text).

DISCUSSION

This study represents a first step toward an objective evaluation of quality of life of CAPD patients with diabetes. A group of non-diabetic CAPD patients was studied in order to highlight quality of life issues relating specifically to the presence of diabetes in CAPD. In addition comparisons were made with non-dialysis diabetic patients (transplant, chronic kidney disease patients).

The response rate of the patients invited to take part in the study was lower than hoped but comparable to response rates seen in similar studies utilizing self-completed questionnaires.[23] This is one of the main drawbacks of such a mode of administration of QoL instruments. In any case, there were no indications that the non-responders had a very different QoL in comparison with the responders, given the similarities of their clinical status, although one cannot exclude possible differences in their psychological or mental status. Those patients with chronic kidney disease and diabetes required clarification of some questions that dealt specifically with dialysis. This perhaps is not surprising, as this instrument (RQLP) was designed primarily for dialysis patients. Alternatively, it may be that patients with early diabetic nephropathy do not want to contemplate the possibility that their condition could result in a future need for dialysis. After occasional clarification, all questions were completed satisfactorily.

The majority of renal transplant patients who responded were very willing and keen to complete both questionnaires, possibly because of past dialysis experience and recognition that their current quality of life was significantly better. In almost all quality of life domains, they reported the best scores despite the fact that their reported co-morbidity and bodily pain was similar to patients with diabetes and chronic kidney disease.

The CAPD: no DM group reported a longer duration of renal replacement therapy, possibly because of the longer survival of non-diabetic CAPD patients on this method. Despite this longer duration in dialysis, they still reported better quality of life than CAPD patients with diabetes.

Metabolic control of diabetic patients as measured by HbA1C was not good. Tzamaloukas et al. have shown previously that good metabolic control of diabetes is associated with better outcome and better quality of life. Therefore, it may be that poor control in these patients may in part be linked to some of the quality of life scores found in the study.[24] Peripheral neuropathy was less severe in transplant patients, a finding that might be related to improvement after transplantation or selection criteria for renal transplantation. CAPD patients with diabetes had a higher, but not statistically significant, incidence of ischemic heart disease, peripheral vascular disease, lower levels of serum albumin, hemoglobin and creatinine, and more severe neuropathy. These factors influenced their quality of life, and after adjustment for these variables, statistical differences between the four groups disappeared. Numerous studies have enabled the identification of the effects of factors such as anaemia, age, co-morbidity, poor nutrition, cardiovascular disease, and others, which can predict worse outcomes for patients with ESRD, especially those with diabetes mellitus.[25],[26] Pain and pruritus were the main physical symptoms affecting quality of life and were better measured by the RQLP questionnaire.

The SF-36 questionnaire produced similar overall perceptions of quality of life in the four groups, and no change in their health status during the previous four weeks, suggesting a stability in their condition. However, the RQLP instrument clearly demonstrated that patients on CAPD were reporting that they were not leading a normal life. This was not evident from the SF-36 instrument. However, quality of life and perceptions of the effects of illness were not associated with functional ability. That is, these patients had a very good adaptation to their clinical status and were leading a satisfactory quality of life. These findings emphasize the difficulties in making decisions about the discontinuation of dialysis treatment in such vulnerable groups of patients, given that many times in clinical settings, diabetic patients exhibit severe comorbidity but are still pleased to be alive and participate in every day's activities.[27]

In general, in both questionnaires, CAPD patients with diabetes had the worst quality of life, with renal transplant patients reporting the best. This finding is in agreement with other studies, leading to the analysis of transplantation as the most effective form of renal replacement therapy.[28] It is of note that it was mainly the physical component of quality of life measurements that produced such bad scoring results of CAPD patients with diabetes, while the mental component had a less dismal influence on their QoL. With the use of both instruments, it was clear in all groups that there was a good mental adaptation to their situation. In addition, with the use of RQLP instrument, it was clear that all patients had a positive attitude for the future, a satisfactory self-esteem, with less preoccupation for sexual and family matters—a finding that enhances the previous thoughts on functional ability and quality of life matters and also shows that although life was not considered normal in this group of patients, it was worthwhile. Both questionnaires appeared able to evaluate quality of life in the four groups, but with the renal-specific instrument providing better discrimination of items that particularly influenced overall quality of life in these patient groups, particular the influence of replacement therapy. Psychosocial aspects of quality of life highlight this. The limitation of social function appeared to be particularly important for both groups of CAPD patients, demonstrating the major influence of the treatment itself and the environment, while, as was said before, financial, sexual, and other matters did not appear significant for their quality of life. In other non-renal patient groups, these domains are usually reported to be important aspects of life quality. These differences are in concordance with the modern perceptions of quality of life. Because of its multidimensional nature and unique personal perception, it is clear that an individual patient's opinion and viewpoint may differ significantly from one group to another. Understanding these differences can give useful and practical information for improvement of their quality of life.[1],[12],[29] In the excellent review article in JAMA[12] evaluating how well QoL is being measured in the medical literature and how a new approach to its measurement can be offered, the authors conclude:

The challenges arise because quality of life, rather than being a mere rating of health status, is actually uniquely personal perception. . . . Accordingly, QoL can be suitably measured only by determining the preferences of patients and supplementing (or replacing) the authoritative opinions of the “experts” (p. 623).

In summary, the QoL of these diabetic CAPD patients was not good, mainly regarding the physical limitations due to many co-morbidity factors. Despite this, they still had a mental health status comparable to the other groups, allowing that their life is worthwhile. In addition, correction for co-morbidity made the differences between groups disappear, something that leads us to apply preventive measures rather than delayed treatment to diabetic complications.

Both instruments were useful for overall assessment of quality of life in CAPD patients with diabetes and were somewhat complementary to each other. However, in an individual case, the RQLP instrument allows much clearer identification of individual factors that have influenced the final score.

It is now evident that health policy makers and medical staff have recognized the importance of measuring quality of life provided by organ replacement treatment. Understanding what influences a patient's life permits healthcare professionals to work toward improving both quantity and quality, such that patients may not in the future need to choose between them. However, it can be detected from this study, and from clinical experience as well, that the correction of accompanying co-morbid factors, especially before commencing dialysis, may offer improvement in quality of life. In addition, it is the renal transplantation that is likely to improve QoL, irrespective of the complication state of the patients.

Notes

18. Salek SM. Development, validation and clinical evaluation of a health-related quality of life instrument. Thesis presented to the University of Wales.