RELATIONSHIP BETWEEN MORTALITY AND TYPE OF VASCULAR ACCESS IN PATIENTS WITH CHRONIC KIDNEY DISEASE UNDERGOING RENAL REPLACEMENT THERAPY WITH HEMODIALYSIS AT A SECONDARY HOSPITAL IN MEXICO.

Chronic kidney disease (CKD) represents a significant public health problem due to its high morbidity and mortality rates, the high costs it generates, and the reduction in quality of life it causes. This pathology is a multifactorial process resulting from multiple chronic degenerative diseases, such as diabetes and systemic arterial hypertension, which are in turn influenced by unhealthy lifestyles and the age of the population. It is progressive and irreversible, generally leading to end-stage disease with impaired renal function, which necessitates renal replacement therapy (RRT) or results in the death of the patient.

Approximately 70% of patients with CKD worldwide urgently require some form of RRT, and hemodialysis (HD) is usually the preferred modality due to the urgency and life-threatening nature of the condition; it is more common in low- and middle-income countries. The United States Renal Data System ranks Mexico second and seventh in incidence and prevalence, respectively, worldwide; The factors that explain the trend toward increased incidence in Mexico and the rest of the world are the aging population and the prevalence of chronic degenerative diseases. It is reported that 40% of people with CKD have diabetes mellitus, 32% have systemic hypertension, and 40% have cardiovascular disease. The most common cause of CKD in Mexico is diabetic nephropathy, followed by high blood pressure and glomerular diseases. Although we do not have a patient registry in our country, the incidence is estimated at 377 cases per million inhabitants and the prevalence at 1,142 cases per million inhabitants; of these, approximately 52,000 patients receive some form of renal support therapy and 82% are patients of the Mexican Social Security Institute (IMSS).

CKD is initially asymptomatic, and even in the terminal stage, clinical manifestations may be minimal and nonspecific, which is why most patients receive care for the first time in emergency situations. In all these cases, RRT is the only option for stabilizing the patient and reducing morbidity and mortality. Among the reasons why a patient may require urgent dialysis is little or no prior education about CKD. It has been estimated that nearly 33% of patients received little or no nephrological care prior to diagnosis. 

Raising awareness about timely RRT is a challenge for physicians, who must help patients choose the appropriate renal replacement therapy modality and prepare dialysis access in a timely manner. The onset of uremic symptoms, poor nutritional status, difficult-to-control hypertension, volume overload, and severe acid-base or electrolyte disorders are considered indications for initiating RRT. However, for some patients, acute pulmonary edema or electrolyte disorders manifest before uremic symptoms, leading to the initiation of emergency dialysis.

Currently, due to the increase in this pathology, healthcare professionals are faced with the challenge of finding an adequate and effective RRT that does not cause further complications or abruptly increase patient morbidity and mortality. Preferring one RRT over another is of utmost importance since, for example, patients who start dialysis therapy through hemodialysis (HD) with a central venous catheter have higher morbidity and mortality, longer hospital stays, and higher healthcare costs compared to those who start peritoneal dialysis (PD) or HD with planned vascular access.

In emergency dialysis, the choice of optimal RRT remains a controversial and researched topic, with continuous RRT and intermittent HD being the most commonly used. However, recent studies have shown that PD offers greater benefits; but a higher probability of associated complications has been found, such as a higher rate of peritonitis, exit site infections, catheter dysfunction due to various causes, and even infusion fluid leakage.

HD is an extracorporeal blood purification technique that partially replaces the kidneys' functions of excreting water and solutes and regulating acid-base and electrolyte balance. It dates back to the 1850s, but it was not until 1925 that George Haas performed the first dialysis on a man. It does not replace endocrine or renal metabolic functions. In many cases, this therapy is initiated through a CVC, which is associated with poor outcomes and high patient care costs.

The need for vascular access for HD is as old as the therapy itself, since access to the bloodstream is essential in order to deliver blood to a dialysis circuit.  The choice of vascular access is fundamental, as it is a key factor in receiving high-quality dialysis therapy and an important part of morbidity and mortality. Today, the vascular accesses and catheters used have improved and continue to evolve to reduce complications.

CKD is in itself an independent risk factor for cardiovascular morbidity, partly because of factors such as hypertension, diabetes, dyslipidemia, and increasing age, but also because of specific factors such as calcium phosphate disorders with arterial and cardiac calcification.

Studies have analyzed the conditions of HD patients, including vascular access. The DOPPS study, in its various phases, describes a progressive increase in the use of CVCs. The ANSWER8 study in 2006 reported high percentages of CVC use in HD patients (41%). In 2009, Gruss et al. published a prospective study of 260 HD patients, reporting up to 47% of patients with CVCs and also describing higher mortality associated with the use of this vascular access, which increased with the duration of use. Various guidelines on vascular access management recommend minimizing the use of CVCs, as this type of access is associated with higher morbidity and mortality. This increased risk of mortality associated with the catheter does not depend solely on the catheter itself, but also on the fact that patients have a poor cardiovascular status, generally related to their advanced age and greater comorbidity, which leads to higher mortality.

Study design and patient population

Our study was an analytical, observational, cross-sectional, retrospective study conducted at a secondary care hospital in Mexico, where data were collected from the medical records of adult patients with stage 5 chronic kidney disease undergoing hemodialysis renal replacement therapy.

The study was approved by the Ethics and Research Committee 17018 with registration number R-2025-1701-028.  The inclusion criteria were records of men and women with stage 5 chronic kidney disease undergoing hemodialysis renal replacement therapy and aged ≥18 years, while the exclusion criteria were records with no record of the type of vascular access and incomplete clinical records.

The current observational study was conducted to assess the relationship between mortality and the type of vascular access in patients with chronic kidney disease undergoing hemodialysis renal replacement therapy at a secondary care center.

Data collection

Follow-up data were collected between January 1, 2024, and June 30, 2024, including clinical and sociodemographic data at the start of replacement therapy and at the end of the follow-up period.

Death events were obtained from death certificates, recording the primary cause of death and taking the date of the last hemodialysis session to determine the time from initiation to death during replacement therapy.

Variables

Our dependent variable was mortality, which we defined as the cessation of vital functions; while the independent variable was the type of vascular access, which we defined as the anatomical point through which renal replacement therapy with hemodialysis is accessed and which was categorized qualitatively, nominally into three main accesses: temporary catheter, permanent catheter, and arteriovenous fistula.

Statistical analysis

In this retrospective study, all analyses were descriptive for the intention-to-treat population. Statistical significance was set at a two-tailed p-value < 0.05. For quantitative variables, measures of central tendency (mean or median) were used with their respective measures of dispersion; for qualitative variables, absolute frequencies and percentages were used.

For qualitative variables, Fisher's exact statistical test was used, and a p-value < 0.05 was taken to establish statistical significance. In the evaluation of the association between vascular access and mortality, a logistic regression model was used to calculate the prevalence risk (PR) with 95% confidence intervals (95% CI), and a p-value < 0.05 was considered to establish statistical significance. Potential confounders included in the model were defined as those variables with a p-value < 0.20 in the bivariate analysis. The results were presented in tables.

The study population was obtained from the database of patients undergoing hemodialysis therapy, from which a total of 414 medical records were identified, of these, 392 met the eligibility criteria, while 21 were excluded;19 because their records contained incomplete information and 3 because they were minors. Details of the selection process results are presented in Figure 1.

Regarding the sociodemographic characteristics of the study population, 44.4% were women and 55.6% were men. The mean age was 53 years with a standard deviation of ±14.6 years, with a median and mode of 57 years; there was wide variability by age group, however, the 61-70 age group predominated with 24.2%. In terms of educational level, basic education predominated with 65.1%; the most common marital status was married with 51.5%; Catholicism was the predominant religion with 73% of the population; the rest of the results are shown in Table 1.

Table 2 shows the clinical characteristics of the study population, where the most relevant finding was that the main etiology of kidney disease was unknown in 88.3% of cases. The predominant vascular access was a permanent catheter in 53.6% of cases, followed by arteriovenous fistula in 25.8% of cases, and finally a temporary catheter in 20.7% of cases. It was found that 68.9% had been on dialysis therapy for >12 months, and the main comorbidities reported were hypertension in 93.4% and type 2 diabetes in 55.4%, with autoimmune diseases being the least frequent at 1.5%.

Of the population studied, nine cases of patients who died during the study period were identified, reporting that five men (55.6%) and four women (44.4%) died, resulting in a mortality rate of 2.3% at the end of the study. The average age was 63.4 years with a standard deviation of ±9.2 years, and the median and mode of this variable was 59 years. Table 3 shows the sociodemographic characteristics of this population, where it was documented that the predominant educational level in the population that died was basic education, at 66.7%.

In terms of marital status, the majority of the deceased population was married (66.7%), and the religion professed by the majority of the deceased population was Catholic (77.8%).

With regard to the clinical characteristics (see Table 4) of the deceased population, it was found that the main etiology of kidney disease was unknown in 77.8%. Seventy-seven point eight percent used temporary vascular access and 22.2% used permanent vascular access; no deceased patients with arteriovenous fistulas were identified.

Regarding the duration of renal replacement therapy with hemodialysis, 77.8% had been on dialysis therapy for 0-3 months. The most frequent comorbidities were hypertension and diabetes, both with a prevalence of 66.7%, followed by cardiovascular diseases with 44.4%. It was documented that 66.7% of the deceased had two or more comorbidities, and two cases (22.2%) were documented without any comorbidities.

The main causes of death were compiled, finding cardiovascular causes in 4 cases (44.4%), infectious causes in 3 cases (33.3%), and metabolic causes in 2 cases (22.2%) (see Table 5).

When performing the bivariate analysis, Fisher's exact test was used to establish the relationship between the study variables. A relationship was demonstrated between the type of vascular access and death, obtaining a p-value of 0.001 (see Table 6).

Poisson regression was performed to observe the clinical factors associated with mortality and the type of vascular access used. In the population with chronic kidney disease undergoing hemodialysis replacement therapy at HGR C/MF No. 1, the prevalence of death in those who had a temporary catheter was 8% (p=0.013) compared to those with a permanent catheter, which was statistically significant.

Similarly, it was documented that there was a comorbidity associated with mortality, which was heart disease, with a prevalence of 7% (p=0.049), which was statistically significant. When performing this regression, some protective factors were found, such as having diabetic nephropathy as the etiology of kidney disease (p=0.029) and some glomerulonephritis (p=0.043), as well as the duration of hemodialysis therapy being 7 to 9 months (p=0.009) and >12 months (p=0.021) for those who showed this behavior (see Table 7).

Similarly, Poisson regression was performed to observe sociodemographic factors, and belonging to the 51-60 age group was found to be a risk factor for mortality, corresponding to a 5% (p=0.026) risk and being statistically significant. As with clinical factors, we found protective factors for mortality, including marital status, with cohabitation providing this protection (p=0.023), as well as Catholic religion (p=0.019), both of which were statistically significant (see Table 8).

STUDY LIMITATIONS

Although the sample size is large, our study is still observational and can therefore only support inferences about associations; confounding factors cannot be ruled out. Ideally, only a randomized clinical trial assigning both access types could clarify whether the relationship between access type and mortality is causal.

We recognize that our protocol also has some limitations, such as the lack of analysis of other risk factors that could impact mortality, some of which could be hemodialysis time, KT/V, ultrafiltration in each hemodialysis session, psychosocial status, and treatment adherence, among others.

FUTURE RESEARCH

For future research, we suggest considering the inclusion of clinical, biochemical, and sociodemographic variables that have been shown to be associated with mortality in hemodialysis patients, as well as extending the study period and evaluating a clinical trial for greater control of variables.

In short, patients undergoing hemodialysis treatment have a high mortality rate, which is associated with factors that can be treated, such as the type of vascular access. At the Mexican Social Security Institute, mortality in hemodialysis associated with the type of vascular access remains below that reported in most studies in other countries, both in Latin America and worldwide.

The association between the type of vascular access and mortality was demonstrated, finding that temporary central venous catheters increase mortality. Therefore, measures will have to be taken regarding the type of access offered to our patients, in addition to raising their awareness of the risks and benefits of all vascular accesses, as well as the types of dialysis available.

The deceased population is characterized mainly by older adults and men, with less than 3 months of hemodialysis treatment, the most frequent comorbidity being high blood pressure. The causes of death are very similar to those reported in national and international studies, with infections and cardiovascular disease ranking first.

For all of the above reasons, we are considering implementing a program for the early detection of kidney disease at primary care facilities, with the aim of halting the progression of the disease and delaying the start of replacement therapy as a matter of urgency, thereby offering better options that are tailored to each patient according to their comorbidities and risk factors.