Vascular access for haemodialysis in Ghana: a single centre experience

¹Department of Surgery, School of Medicine and Dentistry, College of Health Sciences, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana, ²Department of Medicine, School of Medicine and Dentistry, College of Health Sciences Kwame Nkrumah University of Science and Technology and Komfo Anokye Teaching Hospital, Kumasi, Ghana, ³Department of Medical Laboratory Science, School of Allied Health Sciences, University of Cape Coast, Cape Coast, Ghana, ⁴Department of Pharmacy Practice, School of Pharmacy, College of Health Sciences, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana, ⁵Department of Obstetrics and Gynecology, College of Medicine, Pan African University, Institute of Life and Earth Sciences, University of Ibadan, Ibadan, Nigeria, ⁶Department of Medical Laboratory Science, School of Allied Health Sciences, University of Health and Allied Sciences, Ho, Ghana, ⁷Trauma and Specialist Hospital, Ghana Health Service, Winneba, Ghana

^&Corresponding author
Isaac Okyere, Department of Surgery, School of Medicine and Dentistry, College of Health Sciences, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana

Introduction    

Haemodialysis has resulted in increased life expectancy in many patients with chronic kidney disease (CKD) and end stage renal disease (ESRD). Vascular access is the bedrock of haemodialysis. Therefore, an appropriate and adequate vascular access (VA) and its maintenance is of optimum priority in haemodialysis patients. The two major types of vascular access for haemodialysis are the temporary and the permanent vascular access. The temporary vascular access involves the insertion of a venous catheter into one of the central veins of the neck, chest or the thigh, that is the internal jugular, subclavian and the femoral veins for haemodialysis. There are three major types of permanent vascular access. The first involves the insertion of a venous catheter into the central veins of the neck, chest or thigh which is then buried under the skin. This includes a tunnel venous catheter insertion into the central veins of the neck, chest and thigh which is then buried under the skin, next, is the creation of an arteriovenous fistula using the native vessels of the upper limb usually and lastly the creation of an arteriovenous shunt between the native vessels of the upper limb or the thigh using artificial interpositional grafts such as Gore-Tex [1]. The temporary venous access types are used mostly in emergency situations when the patient presents in acute pulmonary oedema necessitating urgent haemodialysis such as in acute kidney injury (AKI), drug overdose or poisoning and ESRD patients who present late. Permanent venous access types are indicated in chronic haemodialysis patients especially patients with ERSD.

The arteriovenous fistula (AVF) is considered the “gold standard” for haemodialysis. AVF is associated with fewer complications such as thrombosis and infection. Moreover, it has longer patency with no immune challenge. Arteriovenous graft (AVG) is considered second best to AVF and is used in patients who have small or inaccessible veins. However, it is associated with higher rates of clots formation, infections and immunological challenges. The central venous catheters (CVCs) which are the temporary venous access are used in emergency situations and are liable to infections, clot formation and decreased durability [2,3]. The type of venous access used for haemodialysis has great impact on patient outcome, for there are associated complications for all types of venous access with the potential effects on the quality of life of the patient. Venous access complications like clot formation, infections and narrowing or stenosis of the central veins are major causes of death in haemodialysis patients [2,4,5]. Due to this, there are recommended guidelines for the use of venous access in different groups of haemodialysis patients[6]. Therefore, to improve the quality of life of haemodialysis patients, there is the need to have effective monitoring and to ensure the appropriate venous access creation and care. In the United States and United Kingdom, there are data on the venous access types used in haemodialysis patients [7,8]. On the contrary, to the best of our knowledge, there are currently no data in Ghana on the types of vascular access used in haemodialysis patients. Therefore, the motivation to conduct a retrospective observational cohort analysis to evaluate the kinds of vascular access currently being used by the different classes of haemodialysis patients in our country with data from the Komfo Anokye Teaching Hospital´s Renal Clinic.

Methods     

Study site/design: a cross sectional retrospective cohort study of patients at the nephrology clinic of the Komfo Anokye Teaching Hospital (KATH) from October 2016 to October 2019. KATH is the second largest hospital in Ghana with a 1000 bed capacity and runs a number of specialized clinics including the renal clinic.

Eligibility criteria: records of patients on haemodialysis within the period of study was used. Renal disease patients who had not begun haemodialysis at the time and those with incomplete data were excluded.

Collection of retrospective data: from the records of the nephrology clinic, we manually collected the data of 80 patients (including males and females) on haemodialysis from October 2016 to October 2019. The data collected included demographics including gender, age, geographical location, marital status, educational level and occupation and clinical data comprising, type and stage of renal disease presented, duration of haemodialysis, type of vascular access and site of venous access.

Ethical consideration: we obtained ethical clearance for the study from the Ethics Committee of Komfo Anokye Teaching Hospital.

Data analysis: data was analyzed with Social Package of Statistical Sciences (SPSS) software version 22.0 (IBM Corp. Armonk, NY, USA). Sociodemographic and clinical variables were presented as frequency and percentages. Cross-tabulations were used to display a breakdown of the categorical data and chi square test was used to test for association between the independent variables (age, gender, marital status, education, employment, location and duration of dialysis) and the dependent variable (vascular access type).

Results     

Sociodemographic characteristics of patients: Table 1 shows the general characteristics of the studied group. This study looked at eighty patients. The mean age was 46 ± 16.0 years with a range of 16-77 years. The highest proportion of the patients were males 52 (65%) and married 63(87.8%). With regards to the level of education and employment status, most of the patients had basic education 29 (36.3%) and were engaged in trading 26 (32.5%). Majority of the patients 47 (58.8%) were living in rural setting. Over half 53 (66.3%) of the patients were diagnosed with end-stage renal disease whereas only 4 (5.0%) had acute kidney injury. For the duration of dialysis, 29 (36.3%) of the patients were on haemodialysis for less than 3 months and this was followed closely by 25 (31.3%) patients who had been on dialysis for over a year.

Distribution of vascular access types: Table 2 shows the distribution of the vascular types, the site of vascular access and the duration of dialysis for the patient. It reports that temporary vascular access was the commonest access type, accounting for 43 (53.8%) of patients, accompanied by arteriovenous fistula 22 (27.5%) with tunneled venous access being the least 15 (18.8%). For the dominant venous access, 65 (81.3%) of the patients had the catheter inserted in the internal jugular vein with one person having the catheter in the subclavian vein (1.3%). Fourteen (15.5%) had their venous catheter in the femoral vein. Majority 49 (61.3%) of the patients had the catheters inserted on the right side of the body either the neck or the thigh with 34 (42.5%) having the catheter for less than 3 months.

Distribution of types of access according to sociodemographic characteristics: Table 3 shows the distribution of types of vascular access according to sociodemographic characteristics such as age, gender, marital status, education, employment and place of abode, whether rural or urban. Temporary vascular access (TVA) is considered as the commonest dialysis access in patients living in rural settings 32/43(74.4%) compared with urban dwellers 11/43(25.6%). Similarly, proportion of patients on tunneled venous access (TUVS) was significantly higher in the rural inhabitants 12/22(54.5%) compared with urban dwellers 10/22(45.5%). On the other hand, arteriovenous fistula (AVF) was relatively lower among rural dwellers 3/15(20%) relative to the urban dwellers 12/15(80%). There was statistically significant association between the type of dialysis access used with area of residence while age, gender, marital status, level of education and employment status were not significantly associated with the type of dialysis access diabetic.

Distribution of types of vascular access according to clinical characteristics: Table 4 reported of the relationship between vascular access types and the clinical characteristics of the studied patients. End-stage renal disease patients were the majority of patients with temporary vascular access 22/43 (51.2%) and tunneled venous access 16/22 (72.7%). All the patients with arteriovenous fistula access for haemodialysis were end-stage renal disease patients. For the duration on dialysis 25/43 (58.1%) of the patients with temporary venous access catheters had been on dialysis for less than 3 months and for patients who had been on dialysis for more than a year, 13/15 (86.7%) of them were using an arteriovenous fistula access, with 9/22 (40.9%) having tunneled venous access catheters. It was observed that diagnosis and duration of dialysis were statistically significantly associated with the type of dialysis vascular access.

Discussion     

This study evaluated the types of vascular access used by patients on haemodialysis at the nephrology unit of the Komfo Anokye Teaching Hospital with consideration of their sociodemographic and clinical characteristics. We studied a total of 80 patients with 65% of them being males, over 50% being more than 45 years and married patients forming 87.8%. Majority (59.2%) of the patients were rural dwellers with 32.5% being traders and 27.5% being unemployed. End stage renal disease (ESRD) was the leading disease for the patients on dialysis in our centre constituting 65% of all the renal diseases. This was followed by chronic kidney disease with Acute Kidney Injury (CKD+AKI) (28.8%) and acute kidney injury (AKI) alone forming about AKI (5.1%). Among the 35.5% who had been on dialysis for less than 3 months, 85.2% were using temporary venous access catheters for haemodialysis with the rest, 14.8% using tunneled venous catheters. However, of the 32.9% who had been on dialysis for over a year, 12% had temporary venous access, 52% were using arteriovenous fistula (AVF) and 36% having tunneled venous catheters (TUV). The dominant vascular access type in our centre was therefore temporary venous access constituting 76.9% among rural dwellers but the leading vascular access among the urban dwellers was arteriovenous fistula (AVF) 80.0%. The commonest venous access for catheterization was the internal jugular vein in 80.5% of the patients. All the AKI and majority of the AKI/CKD patients (73.9%) had temporary venous access. Most (51.2%) of the ESRD patients had temporary venous access at the initiation of dialysis. Although there was no significant association between demographics and the type of vascular access, we found a significant association between diagnosis as well as duration of dialysis and the type of vascular access.

Literature has observed that males usually begin haemodialysis earlier than females with a slightly younger average age of onset according to Lee et al. in 2017 and Weigert et al. in 2019 [9, 10]. Similar to the findings of Lee et al. and Ekpe & Ekirikpo in 2010 [9,11], we also observed a preponderance of male patients on haemodialysis. We had 52 males and 28 females on haemodialysis in our study. Though in developed countries with higher life expectancy and good health systems, dialysis is found in more elderly people, except patients who present with AKI or drug poisoning or intoxication which has no age preference and the fact that stage 5 chronic kidney disease (CKD) mostly occurs in old age [12]. We had a relatively younger age for our distribution with the mean age being over 45 years. The youngest patient in our cohort was 16 years. The internal jugular vein approach is the most frequently used vein for catheterization because of its low complication rate, ease of control of haematoma during insertion and the ease of cannulation [13]. Hence it is not surprising that the internal jugular vein was the dominant (80.5%) site for catheterization in our patients. Arteriovenous fistula (AVF) vascular access is the best vascular access for haemodialysis from evidence-based analysis (national kidney foundation 2006) and as such, is expected to be the most common vascular access type used in chronic haemodialysis patients. This however, was not the observation in our study. Only about 28.3% of the study cohort were using arteriovenous fistula for haemodialysis, however, this is similar in the USA where according to the United States Renal Data System, less than 20% of patients used arteriovenous fistula at the onset of dialysis. Although this contradicts with the Kidney Disease Outcomes Quality Initiative (KDOQI) target of at least 50% of haemodialysis patients using AVF at the start of dialysis. However, a higher use of arteriovenous fistula vascular access for haemodialysis has been found in other centres in the world such as 69.3% in Palestine, 77% in France, 90% in Italy, 84% in Germany, 67% in the UK, and 67.3% in Egypt [14,15].

In line with the study of Roca-Tey in 2015, [16], we also observed that about 86.7% of those who had been on dialysis for over a year had arteriovenous fistula. Due to their prolonged duration on dialysis, nephrologists find it more ideal to introduce AVFs to these patients along the line because of its durability, cost and safety as compared to the rest of the vascular access types. Also, most patients who initiate haemodialysis with TVA and TUV usually switch to AVFs during the first year of dialysis after AVF has fully matured. Therefore, the general expectation is that most, if not all, chronic hemodialysis patients should have arteriovenous fistula. However, about 48% of those who had been on dialysis for over a year still had TVA and TUV in our cohort study. Considering the economic standing of our patients as shown in Table 3, we could attribute this to the inability for most of the patients to afford the cost of the creation of the arteriovenous fistula by the vascular surgeon. In addition, there is also higher rate of arteriovenous failure especially in diabetic patients with diabetic nephropathy [17]. The least (18.8%) used vascular access in our study was tunneled venous access as presented in Table 2. This is comparable to the study of Roca-Tey, where tunneled central catheters were the second least vascular access type used after arteriovenous graft and this form of vascular access is associated with more complications and is mostly only indicated in emergencies and conditions where haemodialysis is temporary [16]. Unlike Roca-Tey et al. none of our patients had arteriovenous graft. Also, arteriovenous graft is considered second best to arteriovenous fistula according to Stehman-Breen et al. [18], hence both patients and clinicians would likely prefer AVF to AVG if available.

Ultimately, in accordance to the findings of Ekpe & Ekirikpo, Rich et al. Kabinga et al. and Hamdan et al. [11, 19,20,14], we found that the use of temporary vascular access in our patients was significantly higher (53.8%) than the permanent vascular access. The late referral or no referral at all to both nephrologists and vascular surgeons, ignorance, cost and unavailability of synthetic grafts, cost of procedure, and patient refusal to undertake the procedures are probable reasons [11]. Unfortunately, the treatment and management of CKD and ESRD is costly worldwide [20] and is excluded from the country´s National Health Insurance Scheme (NHIS) making optimum patient care difficult since all the financial burden falls upon the patients or relatives or benevolent groups. This explains why more of the urban dwellers had AVF while rural dwellers had TVA, as urban dwellers are considered more financially capable with better educational background with clear understanding of the disease process and may have the ability to afford the procedure. Urban dwellers also have better access since most of the hospital facilities are located in the cities.

Conclusion     

The leading vascular access type for patients on haemodialysis in our centre is temporary venous access contrary to world-wide recommendation which should be arteriovenous fistula. Therefore, early referral of CKD and ESRD patients for permanent vascular access procedures should be encouraged. Treatment and management of CKD and ESRD should be included in the nation´s national health insurance to encourage the financially deprived patients to seek and enjoy quality care.

Competing interests     

The authors declare no competing interest.

Authors' contributions     

IO, PO, RKDE, and KAH conceived the concept. IO, PO, RKDE, DS, AAY, PA and GE did the data collection and data analysis. IO, GE, RKDE, PO, and PA did the literature search and prepared the first draft. All the authors have read and agreed to the final manuscript.

Acknowledgments     

We acknowledge the immense role of the staff of the records department in making this work a success.

Tables     

Table 1: sociodemographic characteristics of patients

Table 2: distribution of vascular access types

Table 3: distribution of types of access according to sociodemographic characteristics

Table 4: distribution of types of vascular access according to clinical characteristics

References