Abstract
Anticipatory care planning in the renal unit
Renal Unit, Raigmore Hospital, Inverness, Scotland
Patients with chronic illnesses should have the opportunity to discuss future plans and wishes regarding end-of-life care. In the renal unit at Raigmore Hospital (excluding satellite units), we have been offering the opportunity for anticipatory care planning (ACP) for haemodialysis (HD) patients since April 2011. A renal general practitioner and renal nurse with Consultant support have spent time offering this service. A screening tool is used to identify patients who should be offered ACP. A patient symptom review leads into full ACP discussions. We present data on ACPs and deaths in the HD population between April 2011 and June 2013. Thirty patients have had an ACP in place; four patients approached have declined to have further discussions. This represents 26% of our prevalent HD patients between April 2011 and June 2013. Currently, 12 patients have an ACP in place representing 22% of our prevalent patients. There were 29 deaths (18 male, 11 female; mean age 67.8) in the 27-month period. Twenty-four patients had more than five comorbidities with only five patients having less than five comorbidities. The median time on dialysis for these 29 patients was 71 months (range 3–312 months). An ACP was in place for 18 patients, of these 16 had a DNA CPR order. There were two ACPs which specifically requested CPR and a further two patients who had no ACP but had a DNA CPR order. The mean age for patients with an ACP was 69.2 years. The mean age of the 11 patients who died without an ACP was 64.5 years. The causes of death were cardiac arrest (n = 8), planned withdrawal (n = 8), myocardial infarction (MI; n = 5), pneumonia (n = 2), post surgery (n = 2), cardiac failure (n = 1), haemorrhage from fistula (n = 1), metastatic carcinoma (n = 1) and IHD (n = 1). Seventy-two percent of deaths were expected (of these 71% had an ACP in place). Places of death were hospital (n = 14), home (n = 9), public place (n = 3), hospice (n = 2) and nursing home (n = 1). Withdrawal of dialysis occurred in eight patients (five died at home, two in a hospice and one in a hospital), seven of these patients had an ACP. Of the unexpected deaths (n = 8), three had a cardiac arrest in a public place, two died post access surgery, one died from haemorrhage from fistula at home and two died from presumed MI at home. Six of these patients had no ACP in place, and 83.5% of patients who had an ACP in place had their wishes upheld. ACP is a time-consuming but rewarding process. We are trying to identify patients who we think may be in their final year of life. Sixty-two percent of HD patients who died between April 2011 and June 2013 had an ACP in place.
Areas of clinical concern which contributed to or caused death in patients on RRT: a review of the Scottish Mortality Audit of Renal Replacement Therapy (SMARRT) data 2008–2012
1Glasgow Renal and Transplant Unit, Glasgow, UK
2Technical Director, SRR, Scotland
3Chair SRR - on behalf of the SMARRT steering group, Scotland
Introduction
After every death of a patient receiving renal replacement therapy (RRT) in Scotland, their clinician considers the circumstances of death and records the presence or absence of areas of concern using a five-point scale. This ranges from ‘no areas of concern’ through to ‘areas of concern which caused the death of this patient’. We present here demographic data on all deaths among RRT patients in Scotland between 2008 and 2012. Furthermore, a detailed analysis of those cases where areas of concern in management have contributed to or caused death is undertaken.
Methods
Data on age, modality of RRT, circumstances, cause and location of death were collected via an audit form available to all renal units throughout Scotland and entered retrospectively into the SRR database. Where the management of the patient was categorised as contributing to or causing death, further details of the circumstances of death were obtained. For deaths occurring in 2008–2009, this was undertaken by a group of consultant nephrologists performing a case note review. For deaths occurring in 2010–2012, further details were obtained from reports of morbidity and mortality meetings, critical incident reviews or investigations undertaken by procurators fiscal. From analysis of this additional information, themes were identified. Each theme was further analysed to identify the recurrent issues for descriptive purposes.
Results
A total of 2172 deaths occurred in the period 1 January 2008–31 December 2012. Of these, 1271 were male (58.5%). Median age at death was 71.8 years (18–98). Areas of clinical concern which may have contributed to or caused death were identified in 82 cases (3.8%). Seven themes were identified; hyperkalaemia, prescribing, systems of care, infection, vascular access, complications of intervention and other. Infection made up the largest single group (33%). Hyperkalaemic deaths tended to occur in younger patients (31 vs. 65.5 years). Analysis within each theme revealed that line sepsis accounts for 44% of infection-related concerns. Anticoagulant drugs, opioids and immunosuppressants were collectively responsible for 85% of prescribing-related concerns. More than one-third of the systems of care concerns were caused by delayed diagnoses (37%).
Conclusions
Significant areas of clinical concern are uncommon in this population accounting for <5% of all deaths over a five-year period. The majority of adverse events leading to death which affect patients receiving RRT are the result of commonly occurring problems such as infection, hyperkalaemia and the organisation of care rather than the delivery of RRT itself. Efforts to improve patient safety should focus on these commonly identified areas of harm.3Chair SRR – on behalf of the SMARRT steering group, Scotland
Fracture incidence in patients on renal replacement therapy
1Glasgow Renal and Transplant Unit, Glasgow, UK
2Renal Unit, University Hospital Crosshouse, Scotland
3Renal Unit, Monklands Hospital, Scotland
4Renal Unit, Dumfries and Galloway Royal Infirmary, Scotland
Introduction
Patients on renal replacement therapy (RRT) are at increased risk of bone fracture because of altered bone metabolism termed chronic kidney disease-mineral and bone disorders (CKD-MBD). The true fracture incidence and consequent risk of fracture in prevalent RRT patients and the association with surrogate markers of CKD-MBD is poorly defined.
Aim
The aim was to quantify the incidence of radiologically proven bone fracture in prevalent RRT patients and compare renal transplant (RT) and dialysis patients.
Methods
We undertook a retrospective analysis of electronic patient records for all prevalent RT, haemodialysis (HD) and peritoneal dialysis (PD) patients across the West of Scotland. Our entry point was 7 July 2010 except for patients from Lanarkshire (1 July 2011), and patients were followed until August 2013. All radiology reports from all hospitals in the West of Scotland attached to patients’ records were included and searched to determine the number of fractures per 1000 patient days. The endpoint of follow-up was defined by date of death or last documented biochemistry result.
Results
We identified 2096 patients on RRT at the start of the study. The prevalence of RRT modality at entry was RT 1081 patients (51.6%), HD 907 patients (43.3%) and PD 108 patients (5.2%).The mean age of RT patients was 50.4 ± 15.3 years, HD patients 61.8 ± 15.8 years and PD patients 57.9 ± 15.3 years. Median duration of follow-up for RT patients was 1112 days, range 8–1155; HD patients 1086 days, range 4–1189 and PD patients 1126 days, range 13–1155. Fracture incidence for each RRT modality was RT: 0.10 fractures per 1000 patient days, HD: 0.28 fractures per 1000 patient days and PD: 0.17 per 1000 patient days.
Conclusions
Our study suggests that fracture incidence may be higher in patients on HD compared to RT patients and patients on PD. Further research is warranted to determine if the association between RRT modality and fracture incidence is independent of other risk factors for fracture.
Fosfomycin – is a second dose needed?
Renal Unit, Ninewells Hospital, Dundee, Scotland, UK
Background
Fosfomycin is a broad-spectrum antibiotic commonly used to treat uncomplicated urinary tract infections (UTIs). It is usually administered orally as a single-dose antibiotic. However, it is not clear if there is a difference between single- and multiple-dose treatments in terms of time to recurrence of UTI. Similar uncertainty exists with the use of Fosfomycin in renal transplant population.
Aim
In this single-centre retrospective analysis, we examined the use of Fosfomycin using dispensary records. The aim of this study was to examine the impact of single versus multiple doses of Fosfomycin on the time to recurrence of UTI.
Methods
Data on the use of Fosfomycin over the period from December 2010 to March 2013 were collected from the dispensary of Ninewells Hospital, Dundee, Scotland. Electronic data on age, sex, baseline estimated glomerular filtration rate (eGFR) and type of uropathogens were gathered. Patients were divided, broadly, into those with single versus multiple Fosfomycin doses. The time from treatment to the next positive urine culture was calculated in days for each patient.
Results
A total of 79 patients who received Fosfomycin therapy over the specified period were included in the analysis. Fifty patients had a single dose and 29 patients had multiple doses. There was no significant statistical difference between both groups in the time to recurrence of UTI (p = 0.89). Similarly, we could not demonstrate any advantage of multiple dosing in the renal transplant patients (11 vs. 16, p = 0.6).
Conclusion
There is no significant difference between single versus multiple Fosfomycin doses. Single-dose treatment should be preferred as compliance will be improved whilst reducing adverse effects and costs.
Incidence and outcomes of acute kidney injury requiring renal replacement therapy in Tayside
Renal Unit, Ninewells Hospital, Dundee, Scotland, UK
Introduction
The incidence of acute kidney injury (AKI) requiring dialysis is rising globally and is associated with high mortality and morbidity. KDIGO (Kidney Disease: Improving Global Outcomes) proposed a uniform standard to diagnose and classify AKI in 2007. The aim of this study was to compare the incidence of AKI requiring renal replacement therapy (RRT) during 2012 in Tayside with previous data and determine the impact of RRT on morbidity, mortality and hospital stay.
Methods
We retrospectively examined all patients who underwent RRT in Ninewells Hospital from 1 January 2012 to 31 December 2012. This included patients in the Intensive Care Unit (ICU), medical and surgical high dependency areas and the renal ward. Patients with known end-stage renal disease (ESRD) established on dialysis were excluded. Conventional and electronic patient records were used to collect outcome data at discharge, 90 days and 12 months.
Results
We identified 179 individuals who received RRT during 2012. The incidence of AKI requiring RRT was calculated as 430 per million population per year. Of the patients receiving RRT, 65% were male, 24% was diabetic and the mean age was 67 years (SD
Conclusion
The incidence of AKI requiring RRT in our study is remarkably higher than estimated in previous studies performed in Scotland. AKI requiring RRT is associated with high mortality and impaired renal recovery in our population.
Hospital mortality at the weekend: case-mix or consultant working?
Western Infirmary Renal Unit, Glasgow, UK
Introduction
There are longstanding fundamental variations in National Health Service provision between weekend and weekday periods. This has been a highly publicised topic of political debate with a recent report suggesting a 10% increase in mortality of patients admitted at weekends. This study aims to assess in-patient mortality patterns by day of the week, in a tertiary referral renal unit with established seven-day consultant working, over a two-year observation period.
Methods
All admissions to the Renal Unit at the Western Infirmary Glasgow from 1 August 2011 to 31 July 2013 were identified by searching the Strathclyde Electronic Renal Patient Record database. Mortality related to day of admission and related to day of death was analysed. Data on bed occupancy for each day of the week were collated. The typical caseload of the department was also assessed by detailing the circumstances of each patient’s admission over a concurrent four-week period of observation. Day of admission data. There were 46 deaths within 30 days of admission amongst patients admitted on a weekend day, compared with 107 deaths within 30 days of admission amongst patients admitted on a weekday. This equates to a 30-day mortality incidence of 46/208 = 0.22 deaths per weekend day, compared with 107/523 = 0.20 deaths per weekday. When expressed as a proportion of admissions, 30-day mortality was higher for those admitted on a weekend day 46/540 (8.5%) compared with those on a weekday 107/2013 (5.3%) (RR 1.60, 95% CI 1.15–2.23, p = 0.005). Day of death data. There were 48 deaths within 30 days of admission over 8118 in-patient weekend days (48/8118 = 0.0059 deaths per patient weekend day) compared with 105 deaths within 30 days of admission over 21854 in-patient weekdays (105/21854 = 0.0048 deaths per patient weekday). This provides a relative risk of 1.23 (95% CI 0.88–1.73, p = 0.23) with regard to risk of death on a weekend day. When investigating case-mix, we found that a greater proportion of inter-hospital transfers occurred during weekend days than weekdays. Notably over a 4 week sample of admissions, there were 10 acute kidney injury (AKI) admissions out of eight observed weekend days (rate of 1.25/day) compared with 12 out of 20 observed weekdays (rate of 0.6/day). When expressed as a proportion of admissions, AKI accounted for 10/23 weekend admissions and 12/64 weekday admissions (43.5% and 18.8% respectively).
Discussion
Our results support previous research by indicating higher mortality for patients admitted on a weekend. When considering mortality on a given day of the week, we showed no significant difference in relative risk when comparing in-patients present at the weekend to those present on a weekday. These data are generated in a unit with seven-day consultant working. Numerous variables from staffing, clinical resources and interval since dialysis may influence the mortality rate. Our data suggest that the population of patients admitted at the weekend is of a different demographic to those admitted during the working week. This study suggests that case-mix may be an important variable to consider when performing more detailed investigation on weekend admission mortality rates.
Mycophenolate mofetil therapy in IgA nephropathy: histological changes after treatment
Imperial College Kidney and Transplant Institute, London, UK
Background
Endocapillary hypercellularity independently predicts renal outcome in IgA nephropathy (IgAN).1 Mycophenolate mofetil (MMF) treatment is offered to patients presenting to the Imperial College Kidney and Transplant Institute with IgAN and histological evidence of endocapillary inflammation. Clinical trials of MMF in IgAN have been inconclusive. Evidence of histological improvement following MMF treatment would support its therapeutic use. We therefore reviewed histological changes after MMF therapy in a cohort of IgAN patients.
Method
Eleven IgAN patients with repeat native renal biopsies before and after MMF treatment were identified. Patients were excluded if they had received any other immunosuppression therapy, including corticosteroids. Based on the Oxford classification of IgAN,1 we reviewed histological changes after MMF treatment.
Results
Seven patients (60%) were male. At diagnostic renal biopsy, median age was 42 (range 19–67), serum creatinine was 127umol/L (56–233) and urine protein creatinine ratio (UPCR) was 157 mg/dl (67–224). The median time between biopsies was 29 months (14–41).
Conclusion
MMF treatment is associated with histopathological improvement in IgAN.
Reference
Phosphate has deleterious effects on vascular function
Renal Research Group, University of Glasgow, Glasgow, UK
Background
Elevated serum phosphate is an independent risk factor for cardiovascular disease. Whether this is a direct effect of elevated phosphate or dependent on changes in intracellular calcium or calcium/phosphate product is unknown. We examined the direct effects of phosphate concentration in human resistance vessels and human umbilical vein endothelial cells (HUVECs).
Methods
Surplus adipose tissue was removed from patients with chronic kidney disease (CKD) stage 5 undergoing live donor transplantation and their normal donors. Resistance vessels were dissected and incubated in a physiological saline solution (PSS) with normal (1.18 mM) or high phosphate concentration (2.5 mM) for 16 h, then mounted on a myograph. Vasoconstrictor responses to phenylepherine (PE) and vasorelaxation responses to carbachol and sodium nitroprusside (SNP) were measured. Concentration-response curves were constructed for PE, carbachol and SNP. Area under the curve (AUC) was calculated and comparisons were made using either a t-test or an ANOVA. HUVECs were grown in normal (0.5 mM) and high (3 mM) phosphate medium. eNOS and nitrotyrosine expression were measured by Western blot and intracellular calcium concentration measured by epifluorescence with FURA 2 AM. Gene expression was studied with PCR.
Results
Vessels from patients with and without CKD incubated in high phosphate relax less well to carbachol (p < 0.05). Vessels from patients without CKD relax less well to SNP (p < 0.05); this difference is not seen in vessels from patients with CKD. Expression of total and phospho eNOS was reduced in HUVECs grown in high phosphate whilst nitrotyrosine expression was increased. Calcium concentration was not significantly different between HUVECs grown in high and normal phosphate. Genes involved in cell cycle and growth were upregulated, and expression of the phosphate transporters PiT1 and PiT2 was unchanged. HUVECs express Klotho and the FGF 1 receptor.
Conclusions
Elevated phosphate decreases endothelium-dependent vasodilatation in patients with and without CKD. This may be a marker of endothelial dysfunction, supported by the reduced eNOS protein expression and increased nitrotyrosine expression seen in HUVECs. Elevated phosphate also impairs endothelial independent relaxation in vessels from healthy patients without CKD. These experiments indicate direct effects of elevated phosphate on the NO system, and on vascular function, and support the notion that phosphate has direct effects in uremia.
Sustained phosphate loading impairs endothelial function: a single blind cross over trial
Renal Research Group, University of Glasgow, Glasgow, UK
Background
Serum phosphate is linked with increased cardiovascular risk although the mechanism of action is unclear. The effect of sustained short-term phosphate loading on endothelial function has not previously been studied. This study considers the effect of phosphate loading on endothelial function measured by flow-mediated dilatation (FMD).
Methods
Healthy volunteers attended for a baseline and two subsequent visits. Blood was drawn for measures including bone biochemistry, vitamin D, FGF-23 and klotho. A 24-h urine collection was performed prior to attendance, and analysis included urinary cyclic guanosine monophosphate (cGMP) and FGF-23 concentrations. FMD was recorded. Volunteers were randomised to take lanthanum carbonate (LC) or phosphate sandoz (PS) for two weeks prior to the next visit. After a wash out period, volunteers took the other drug and attended for a final visit. One individual, blinded to the order of drug ingestion, performed and analysed each FMD measure.
Results
Of 19 participants, 12 were female. At baseline, mean age was 42 ± 14 years, estimated glomerular filtration rate (eGFR) 102 ± 10 ml/min, serum phosphate 1.05 ± 0.18mmol/L and fractional excretion of phosphate (FeP) 14.3 ± 3.4%. Median FMD was 8.4% (IQR 6.2–11.6%) post cuff inflation. After PS, there was a trend towards a higher serum phosphate within the normal range. FGF-23 and FeP rose significantly compared to baseline (p = 0.013, p < 0.001). FMD post cuff inflation reduced significantly (3.38% (IQR 2.57–5.26%), p < 0.001). With LC, serum phosphate was unchanged. FeP fell (11.4 ± 4.3, p < 0.001). Post cuff inflation FMD fell (6.6% (IQR 3.4–10.3%), p = 0.033). Randomisation order had no effect. In a regression model, higher FeP was an independent predictor of attenuated post cuff inflation FMD (p = 0.02). Urinary cGMP correlated negatively with serum phosphate (p = 0.003).
Conclusions
This is the first study to demonstrate that sustained phosphate loading impairs endothelial function. The observed deleterious effect on FMD seen with PS may be explained by elevated total body phosphate with resultant elevated intra-cellular phosphate. FeP is likely a surrogate marker of total body phosphate. Urinary cGMP, as a marker of endothelial dysfunction negatively correlates with serum phosphate level. This study supports the hypothesis that phosphate increases cardiovascular risk by impairing endothelial function, possibly via the nitric oxide pathway. Sustained phosphate loading is directly detrimental to the vasculature even when serum phosphate remains within the normal range.
Effects of intravenous ascorbic acid on vascular function and oxidative stress in chronic kidney disease
Institute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, UK
Background
Arterial stiffness is associated with excess premature cardiovascular disease in chronic kidney disease (CKD). Antioxidant therapy may represent a therapeutic intervention to improve vascular function and reduce cardiovascular risk. We studied the effect of intravenous ascorbic acid on arterial stiffness in patients with hypertension (HTN) and CKD.
Methods
We performed a crossover study of administration of intravenous ascorbic acid and normal saline in patients with CKD or HTN and normal renal function. Arterial stiffness and serum markers of oxidative stress were measured at each treatment time point. Arterial stiffness indicated by pulse wave velocity and augmentation index was assessed using the SphygmoCor system. Rate of reactive oxygen species (ROS) production was measured using electron paramagnetic resonance, whilst total antioxidant capacity (TAC) was measured by a colorimetric assay.
Results
Fifteen CKD and 15 HTN patients were recruited (mean age 57 years, BMI 29 kg/m2, blood pressure (BP) 144/89 mmHg). The CKD group had a mean estimated glomerular filtration rate (eGFR) of 28 ml/min/1.73 m2 compared to 95 ml/min/1.73 m2 in the HTN group. There was no significant difference in age, lipid profile, glycaemia or BP between groups; however, use of ACE inhibitors, angiotensin receptor blockers, statins and allopurinol was higher in the CKD group. There was a significant reduction in adjusted augmentation index (Aix) after ascorbic acid in both CKD (26%–16%; p < 0.001) and HTN (23%–18%; p = 0.003), but no significant change in pulse wave velocity, and no significant differences between groups. There was a significant increase in rate of ROS production in both CKD (0.433–0.594; p < 0.001) and HTN (0.361–0.516; p < 0.001) which fell to baseline after 60 min. Similarly, there was a significant increase in TAC in both CKD (0.642–1.173; p < 0.001) and HTN (0.579–1.151; p < 0.001) which had fallen to near baseline values at 60 min.
Conclusions
Ascorbic acid ameliorates arterial stiffness and causes a paradoxical increase in ROS production, in both CKD and HTN. Further study is required to determine the clinical implications and long-term effects of this.
Habitual levels and patterns of physical activity in a Scottish haemodialysis population, and their relationship with markers of function and functional capacity
1Queen Margaret University, Edinburgh, UK
2Edinburgh Royal Infirmary, NHS Lothian, Edinburgh, UK
Background
The benefits of targeted exercise interventions in haemodialysis patients are acknowledged.1 However, less is known about habitual levels and patterns of physical activity, particularly the prevalent Scottish population. In addition, in the general population, there is emerging evidence that levels of sedentary activity should be considered as a distinct entity from levels of physical activity and may be independently important for a number of risk factors.
Aim
The primary aim was to examine levels and patterns of physical activity in a cohort of haemodialysis patients within Scotland. Secondary aims were to explore relationships between physical activity levels and markers of function/functional capacity along with markers of nutritional, clinical status and quality of life. Only data relating to function/functional capacity is presented here.
Methods and outcome measures
In this prospective observational study, all patients currently on dialysis in NHS Lothian, for >3 months and able to give informed consent were eligible to participate. Habitual levels and patterns of physical activity were measured using an activPAL™ accelerometer, which detects and classifies an individual's free-living activity into periods spent sitting/lying, standing, stepping and number of steps taken. Patients were asked to wear the accelerometer for a continuous period of five days. Functional capacity was estimated using the 6-min walk test, and function was estimated using the 60-s sit-to-stand test (total number of sit-to-stands in 60 s), handgrip dynamometry and self-reported function were estimated using the KDQOL-PCS.
Results
Significant positive relationships were observed between the number of steps and 6-min walk, number of sit to stands, KDQOL-PCS, but not handgrip. No significant relationships were observed between time spent sitting/lying and standing with 6-min walk, sit to stand or KDQOL-PCS. However, there was a significant inverse relationship between time spent standing with handgrip and for time spent standing with 6-min walk for men but not women.
Conclusion
The preliminary results suggest that dialysis patients do not meet recommendations for the number of daily steps (10,000) and are sedentary for the majority of each day. Whilst further analysis and exploration of patterns of sedentary behaviour are necessary, this may provide the basis for simple low-cost interventions, focusing on encouraging patients to sit less and walk more with a view to improving functional capacity, function and ultimately quality of life.
Reference
Possible factors influencing levels of physical activity in haemodialysis patients
1Queen Margaret University, Edinburgh, UK
2Edinburgh Royal Infirmary, NHS Lothian, Edinburgh, UK
Background
Lower levels of physical activity in haemodialysis patents compared to age-matched sedentary controls have been frequently reported. However, little is known about the potential factors influencing levels of physical activity, which may be pertinent when designing targeted exercise interventions.
Aim
The aim of this study was to explore levels of self-reported physical activity with levels of cognition, self-efficacy, barriers and motivators and quality of life in a cohort of Scottish haemodialysis patients.
Methods
To obtain a representative sample of the current haemodialysis population in Lothian, all patients, who were able to provide informed consent, were eligible to participate. At the beginning of a dialysis session, the Montreal cognitive assessment (MoCA) was administered along with one consolidated questionnaire (comprising demographics; Stanford self-efficacy; exercise benefits/barriers scale; Stanford brief activity survey; LASA quality of life). Patients were classified into two groups (inactive or active) based on self-reported level of physical activity (derived from the Stanford brief physical activity survey component of the questionnaire). Differences between groups were determined using independent t-tests.
Results
Discussion and conclusion
The initial results of this study suggest those individuals who consider themselves as inactive are older, have lower than normal cognitive scores, have less confidence in their ability to exercise (lower levels of perceived self-efficacy) and a lower perceived quality of life. In addition, inactive patients are less likely to be thinking about becoming active (readiness to change). However, those who were inactive positively perceived the benefits of exercise more than those who were active but had a greater perception of barriers to exercise. Whilst further analysis is required, these initial results suggest that consideration should be given to such factors when encouraging or implementing exercise programmes.
Case of VRSE in Ayrshire and Arran. An audit of the use of vancomycin in haemodialysis patients
Renal Unit, Crosshouse Hospital, UK
Introduction
Vancomycin resistance has been reported in clinical isolates of both coagulase-negative staphylococci and Staphylococcus aureus. The emerging threat of widespread vancomycin resistance poses a serious public health concern given the fact that vancomycin has long been the preferred treatment of antibiotic-resistant Gram-positive organisms. Following the identification of vancomycin-resistant Staphylococcus epidermidis (VRSE) in a haemodialysis patient in Ayrshire and Arran, the decision was taken to audit the use of vancomycin and adherence to the current vancomycin protocol in the renal unit.
Methods
All haemodialysis patients in Ayrshire and Arran who had received IV vancomycin over a six-month period were identified using SERPR. Patients who had received a single dose only were excluded from further analysis. Length of course, indication, vancomycin levels, dose given and compliance with protocol were recorded.
Results
Sixty-one courses of vancomycin were prescribed in 47 patients between 14 January 2013 and 14 August 2013. Ten prescriptions were for single doses only and were therefore excluded from further analysis. A total of 51 (n = 51) courses of vancomycin were analysed. The mean length of treatment was 21.2 days (range 3–120). A total of 282 levels were recorded with mean level 15.1mg/L (range <2–30.8). Four hundred eighty-nine doses of vancomycin were administered with mean dose 742.8mg (range 500–1000). All patients were loaded with 1000 mg. Forty-four of 527 dosing decisions were not made according to protocol, of which 33 were due to lack of levels (not available in time, not sent or sent in incorrect bottle).
Conclusions
The majority of decisions regarding vancomycin dosing were made in accordance with the current protocol. Despite this levels are lower than target with 52% <15mg/L. It is widely accepted that sub-therapeutic levels contribute to resistance primarily by allowing the organism to thicken its cell wall. This highlights the need for a new vancomycin protocol with higher loading doses and higher target levels.
Case report: use of bortezomib to treat chronic antibody-mediated rejection in a live-related donor renal transplant recipient
Renal Unit, Royal Infirmary of Edinburgh, UK
Antibody-mediated rejection (AMR) in renal transplant recipients is associated with deterioration in renal function and graft loss. Treatment with plasmapheresis, intravenous immunoglobulins (IVIG) and steroids has failed to substantially improve graft survival in chronic AMR. Bortezomib is a proteasome inhibitor that has been associated with reversal of histopathological changes and prolonged graft survival in chronic AMR. We report a case of bortezomib treatment in a patient with chronic AMR. Our patient is a 35-year-old male with end-stage renal disease secondary to chronic glomerulonephritis. He underwent a live related donor transplant in 2009 and was immune suppressed with prednisolone, mycophenolate mofetil (MMF) and ciclosporin. Day 8 biopsy was borderline for acute cellular rejection, and he received pulsed intravenous methylprednisolone. He had been at a stable serum creatinine (SCr) of 200 and estimated glomerular filtration rate (eGFR) of 30 ml/min for almost four years when he presented in April 2013 with an acute rise in SCr in the context of sepsis and presumed ciclosporin-mediated toxicity, ciclosporin was withdrawn. He went on to have a renal transplant biopsy in July 2013 which was stained C4d positive, and he was found to have high titres of HLA Class I and II antibodies. Further biopsy in August 2013 showed chronic AMR, and the patient was treated with methylprednisolone, plasmapheresis and IVIG. Repeat biopsy performed eight days later showed ongoing severe chronic AMR. After counselling the patient, bortezomib therapy was commenced. The treatment regime consisted of two courses of intravenous bortezomib on days 1, 4, 8 and 11 two weeks apart supported throughout with plasmapheresis. On repeat biopsy at the end of the treatment course, there was an evidence of ongoing chronic AMR. There was, however, reduced inflammation, and the majority of glomeruli remained viable. C4d staining had decreased. SCr improved from 410 to around 340 during treatment and has remained stable. The HLA DQ antibody titre fell to one-third of initial levels and remained stable. The patient suffered no bortezomib-associated complications. We report on a case of late AMR treated with high-dose steroids, plasmapheresis with IVIG and bortezomib. The patient suffered from no major side effects to the treatment regime and showed early evidence of reduced inflammation and stasis of his chronic AMR on repeat biopsy. Serum HLA antibody titres were seen to fall. This case report shows some early efficacy of bortezomib in treating chronic AMR in a renal transplant recipient.
Combined eculizumab and bortezomib as additional therapies for early severe antibody mediated rejection in a highly sensitised renal transplant recipient
1Department of Renal Medicine, Aberdeen Royal Infirmary, UK
2Department of Histocompatibility and Immunogenetics, Royal Infirmary of Edinburgh, UK
3Department of Renal Medicine, Royal Infirmary of Edinburgh, UK
Introduction
The management of early severe acute antibody mediated rejection (AMR) is controversial. Eculizumab, a humanised anti C5-antibody that inhibits terminal complement and bortezomib, a proteasome inhibitor, have been reported individually as effective treatments for severe AMR refractory to conventional therapy.
Case summary
A 32-year-old woman with stage V chronic kidney disease (CKD) secondary to FSGS received a DCD renal transplant on 17 May 2013. She had received a previous renal transplant aged 15 which failed due to rejection at 18 months. As a result of her previous transplant and pregnancy, she was highly sensitised with a match score of 6, which had resulted in a waiting time of over 5000 days and no previous transplant offers. Pre-transplant crossmatch revealed a DPB1y DSA at 10,000 MFI. B flow crossmatch was positive and CDC crossmatch was negative. Standard triple immunosuppression of tacrolimus, mycophenolate mofetil and prednisolone was supplemented with Campath induction and plasma exchange (PEX) with 100 mg/kg intravenous immunoglobulin (IVIG) pre-transplant and then on alternate days post transplant. There was immediate graft function, and creatinine fell to a nadir of 100 µmol/L before rising to 113 µmol/L at day 5 prompting a transplant biopsy. This demonstrated an acute severe thrombotic microangiopathy with diffuse C4d positivity. These findings, in conjunction with a rise in DPB1y DSA to 23,000 MFI with a de novo class I and class II DSA, (cumulative MFI 48,590) were consistent with severe AMR. Biopsy was complicated by a haemaodynamically significant bleed accompanied by a rise in creatinine to 440 umol/L. She was pulsed with methyl prednisolone, and PEX/IVIG frequency was increased to daily treatments once haemodynamically stable. Eculizumab commenced at day 8 for a total of 12 doses over 18 days. Despite on going PEX/IVIG and eculizumab, her cumulative DSA remained >20,000 (1:5 dilution) and creatinine plateaued at 200 µmol/L. She was given three doses of bortezomib, and PEX/IVIG switched to alternate days. Cumulative DSA fell to <20,000. Serum creatinine peaked at 537 µmol/L at day 10, falling to a nadir of 128 µmol/L after the first cycle of bortezomib. Thrice weekly PEX/IVIG was continued and followed by a second cycle of four doses of bortezomib before discontinuing PEX. Out-patient monitoring during the third month post transplant revealed stable graft function with a creatinine ranging between 110 and 120 µmol/L. Elective biopsy on day 89 demonstrated resolution of the TMA with minimal inflammation though remained diffusely C4d positive. The patient remains under close follow-up with a current creatinine of 100 µmol/L and stable DSA titres.
Discussion
The dual strategy of eculizumab and bortezomib with PEX and IVIG has not been widely reported in the literature. In this case of severe TMA resulting from acute AMR, which would almost certainly have resulted in graft loss, we observed rapid recovery of graft function and a drop in cumulative DSA, permitting withdrawal of PEX with no evidence of relapse or transplant glomerulopathy at three months. The treatment was well tolerated with minimal adverse effects to date.
Renal function, serum phosphate, serum-adjusted calcium and blood pressure following live donor nephrectomy
Glasgow Renal and Transplant Unit, UK
Background/Aim
Live donor nephrectomy is associated with compensatory hyperfiltration in the remaining kidney. In the general population, elevated serum phosphate, within the normal range, is associated with adverse cardiovascular outcomes. The effect of nephrectomy on serum phosphate is not well studied. The aim of this study was to quantify the effect of nephrectomy on glomerular filtration rate (GFR), serum phosphate, serum-adjusted calcium and blood pressure in the first year after donation.
Method
Data were obtained from the electronic patient record. Patients with a recorded procedure of ‘live donor nephrectomy’ between 2005 and 2013 were identified. GFR after donation was calculated using the formula ‘GFR = Preoperative Isotope GFR × (Serum creatinine/Preoperative Serum Creatinine)’, which is accurate assuming muscle mass does not change. Pre and post (three days, two weeks, three months and one year) donation serum creatinine, phosphate and adjusted calcium were recorded. Pre and post donation (three months and one year) blood pressure was recorded.
Results
One hundred and ninety-four consecutive donors were identified, of whom nine had insufficient follow-up data for inclusion. The mean age was 46.1 ± 11.4 years and 47% (n = 90) were male. Mean pre-donation isotope GFR was 97.8 ± 17.1 ml/min/1.73m2. Post donation calculated GFR was significantly lower at 72%, 70%, 71% and 75% of pre-donation GFR at three days, two weeks, three months and one year, respectively (p < 0.001). Post donation serum phosphate was significantly lower at three days post donation compared with pre-donation value (0.72 ± 0.16 vs. 1.04 ± 0.16 mmol/l; p < 0.001). There was no significant difference in serum phosphate at two weeks, three months or one year post donation. Mean pre-donation adjusted serum calcium was 2.40 mmol/l ± 0.08 mmol/l. Post donation adjusted serum calcium was significantly lower at one year compared with pre-donation values. (2.35 ± 0.13 vs. 2.41 ± 0.08 mmol/l; p = 0.005). Mean pre-donation blood pressure was 132/78 ± 16/10 mmHg. At three months, BP was 129/77 mmHg ± 16/9 and at one year 131/79 ± 17/11 mmHg.
Conclusion
Individuals who undergo live donor nephrectomy have a GFR of 75% at one year compared to pre-donation GFR. Compensation in GFR occurs rapidly after nephrectomy. Post-donation serum phosphate and blood pressure are maintained at one year, whereas serum calcium falls significantly. This observation may relate to changes in the homeostatic regulation of calcium and phosphate.
How erythropoietin treatment targets affect quality of life, mortality and the risk of cardiovascular events in chronic kidney disease: a review
University of Aberdeen, Aberdeen, UK
Introduction
Chronic kidney disease (CKD) is a major public health problem with high morbidity and mortality. Anaemia due to reduced synthesis of erythropoietin often develops in those with CKD and can be treated with erythropoietin-stimulating agents. The current NICE guideline (Updated February 2011) recommends sub-normal target haemoglobin (Hb) level (10–12 g/dl) for CKD patients. However, some observational studies suggest benefits of higher Hb target levels.
Aim
To evaluate current evidence as to whether erythropoietin treatment with a high target Hb level (vs. low) is more effective for reducing mortality, cardiovascular events and improving quality of life in pre-dialysis CKD patients.
Methods
Medline and EMBASE literature searches (May 2013) using terms for chronic kidney disease, erythropoietin, anaemia and the outcomes of interest (mortality, cardiovascular disease and quality of life) were combined using Boolean operators ‘AND’ or ‘OR’, as appropriate. Randomised control trials (RCTs) comparing the effects of high Hb targets with low Hb targets in pre-dialysis adult patients were included. Haemodialysis and peritoneal dialysis patients, paediatric studies and studies which were not randomised controlled trials were excluded. The Critical Appraisal Skills Programme (CASP) was used for quality assessment. Scoring from 0 to 9 was given, where 9 being the best possible quality, fulfilling all the quality criteria (precision, blinding, appropriate intervention and comparison of study groups). Two reviewers appraised each study for inclusion or otherwise and extracted appropriate data from the studies. Results were expressed as hazard ratios with 95% confidence intervals (CI).
Results
The search generated 100 citations. Seven passed the inclusion and exclusion criteria. Of these seven studies (6945 individuals in total), only Villar et al. (May 2011) was not included in the NICE guideline review. The quality of the studies was variable. The CASP scores of the seven studies ranged from 5 to 8. Five studies indicated higher all-cause-mortality risk in the high target Hb group; however, the CI all crossed unity. Six studies reported on myocardial infarctions but did not show significant difference between the Hb target groups. Three studies (n = 4962) reporting on stroke noted a higher risk in the high Hb target group, but only one study (n = 4038) reported it as statistically significant with hazard ratio of 1.92 (CI: 1.38–2.68). Five studies assessed quality of life as an outcome; three reported a significant improvement with higher Hb target level.
Conclusion
This review demonstrates that current evidence supports the 2011 NICE guideline recommending sub-normal target level for CKD patients. The quality of evidence is variable. CKD patients with high Hb target level generally report improved quality of life, however, also has increased risk of stroke. Ideally, future studies should look to stratify the risk of cardiovascular events and mortality in CKD patients with cardiovascular co-morbidities as well as the life expectancy in this group of patients to explore whether the increased quality of life outweighs the increased risk of stroke.
The concept of ‘chronic lithium toxicity’: an epidemiological approach
1Department of Biomedical Science and Public Health, Medical Research Institute, University of Dundee, Scotland
2Health Informatics Centre (HIC), University of Dundee, Scotland
3Renal Unit, Ninewells Hospital and Medical School, Dundee, Scotland
4Dundee Epidemiology and Biostatistics Unit, Division of Population Health Sciences, Medical Research Institute, University of Dundee, Scotland
Background
For over 40 years, there has been a debate about the long-term effect of lithium maintenance therapy on renal function. This population record-linkage study set out to assess whether there was evidence of this long-held belief.
Methods
We used datasets from the University of Dundee’s Health Informatics Centre (HIC) to link laboratory data to community prescriptions for Tayside, Scotland. Primary outcome was the estimated glomerular filtration rate (eGFR) using the chronic kidney disease (CKD)-EPI equation. A cohort of patients newly commenced on lithium was constructed between January 2000 and January 2012. Patients with incidence exposure to other first-line drugs (quetiapine, olanzapine and semisodium valproate) provided a natural comparator group. Patients with glomerular/tubulo-interstitial disease or CKD stages 4/5 at baseline were excluded. Analysis of outcome utilised random coefficient models.
Findings
A total of 1120 patients (305 lithium, 815 comparator drugs) aged 18–65 years qualified for inclusion providing 13,963 estimated glomerular filtration rate (eGFR) values over 12 years. Mean lithium exposure length was 55 months (SD 42, min 6, max 144). Mean adjusted annual decline in eGFR was 1.5 ml/min/1.73 m2 (SE 4.2 ml/min/1.73 m2), with no difference between lithium and comparator groups. The final model identified significant predictors for a decline in eGFR as age, baseline eGFR, diabetes, hypertension, co-prescriptions of nephrotoxic drugs (penicillinase-resistant penicillins at baseline; ACE inhibitors, diuretics and NSAIDs during F/U) and episodes of lithium toxicity (>1.5 mmol/L) but not lithium exposure length or mean lithium serum level.
Proceed
Interpretation: The analysis suggests no effect of lithium maintenance therapy (lithium levels in therapeutic range) on the rate of change in eGFR over time. The model, however, needs to be tested in a dataset other than the derivation data. Our results therefore contradict the concept that long-term lithium therapy is associated with nephrotoxicity in the absence of episodes of acute intoxication and that duration of therapy and cumulative dose are the major determinants of toxicity. Previous publications inferring exposure length as being a predictor for renal decline in patients on long-term lithium therapy will have been affected by inappropriate study design and associated bias and confounding. The concept of ‘chronic lithium nephrotoxicity in the absence of episodes of acute intoxication’ may well be a modern medical myth.
Reducing haemodialysis bacteraemia rates
Renal Unit, Western Infirmary, Dumbarton Road, Glasgow, UK
Introduction
Patients on renal replacement therapy (RRT) experience significantly higher rates of morbidity and mortality than the general population. Infection is the second highest cause of death for patients on RRT, accounting for approximately 20% of mortality.1 Up to 90% of bacteraemias in the haemodialysis population are staphylococcal,2 supporting vascular access as the primary source of infection. Preventing infection-related deaths on HD requires two things: (i) ensuring optimal vascular access with an arteriovenous fistula (AVF) or graft (AVG), as opposed to tunnelled or non-tunnelled central venous catheter (TCVC/NTCVC) and (ii) reducing the prevailing bacteraemia rate. In this study, we describe the effect of introducing catheter lock solution taurolidine-citrate heparin (TauroHep500©) in TCVCs and chlorhexidine-impregnated exit-site patch (Biopatch©) in NTCVCs on staphylococcal bloodstream infection rates.
Methods
Data on the quarterly incidence rates of all staphylococcal bacteraemic events in the NHS Greater Glasgow and Clyde and NHS Forth Valley haemodialysis population were collected for the period from April 2011 to June 2013. This was achieved by a structured query language (SQL) interrogation of the renal unit electronic patient record (EPR) with the resulting output being processed by Microsoft Office Excel 2003. Consecutive blood culture results >14 days apart were regarded as separate events. Vascular access data were cross-checked by hand. Event rates were expressed as events per 1000 haemodialysis-exposed days for each vascular access type. Comparison between periods was made by student’s t-testing with a significance level set at α < 0.05.
Results
Two hundred sixty-one bacteraemia events occurred over 427,836 HD days; 79 events in 289,456 AVF/AVG HD days, 160 events in 135,537 TCVC HD days and 22 events in 2834 NTCVC HD days. Comparing the staphylococcal bacteraemia rate before and after the introduction of TauroHep500 in TCVCs demonstrated a reduction from 1.58/1000 HD days (95% confidence interval [CI] 1.12–2.03) to 0.74/1000 HD days (95% CI 0.20–1.28), p < 0.01. In NTCVCs, no corresponding reduction in staphylococcal bacteraemia rates was observed; 6.06/1000 HD days (95% CI 3.43–8.71) pre-Biopatch introduction, 10.36/1000 HD days (95% CI 3.10–17.62) following introduction, p = 0.10. The staphylococcal bacteraemia rate in AVF/AVGs remained unchanged; 0.28/1000HD days pre-July 2012 (95% CI 0.16–0.41) versus 0.27/1000 HD days post-July 2012 (95% CI 0.18–0.36), p = 0.87.
Conclusion
Using chlorhexidine-impregnated foam patches on NTCVC exit sites failed to demonstrate any reduction in bacteraemia rates, though cumulative NTCVC HD days and incidence events were too low to draw any conclusion reliably. Replacing heparin 5000 IU/mL with TauroHep500 as catheter lock solution in patients with TCVCs was associated with a statistically significant 47% reduction in staphylococcal bloodstream infection rates.
References
Posterior reversible encephalopathy syndrome associated with hypertensive hypokalaemic hyporeninaemic hypoaldosteronism
1Department of Cardiology, Forth Valley Royal Hospital, UK
2Renal Unit, Ninewells Hospital and Medical School, Dundee, UK
Posterior reversible encephalopathy syndrome (PRES) is a clinical entity consisting reversible neurological clinical and radiological signs associated with cerebral vasogenic oedema and vascular endothelial dysfunction. The condition can occur in the context of systemic hypertension, eclampsia, sepsis, renal failure, connective tissue disorders, immunosuppression in solid organ transplant recipients and chemotherapeutic agents. We describe an interesting case of PRES in the setting of secondary hypokalaemic hypertension. Our patient is a 29-year-old male, previously well except for mild asthma, who initially presented to the department of urology in our hospital with left-sided abdominal pain, vomiting, haemoproteinuria and an acute kidney injury with a creatinine of 268 µmol/L. The urine protein/creatinine ratio was 67 mg/mmol. He was noted to be hypokalaemic at 3.1 mmol/L, in the absence of other electrolyte abnormalities. The patient’s transtubular potassium gradient was inappropriately raised, with correlating low plasma and aldosterone concentrations. There was no clinical or biochemical evidence to suspect an underlying infective process. A computed tomography (CT) urogram showed normal-sized kidneys with good parenchymal width with no evidence of obstruction or renal calculi. Subsequently, he developed a sudden onset severe fronto-occipital headache associated with an acute visual loss and a blood pressure of 174/99 mmHg. Bedside fundoscopy was unremarkable. Emergency hypertension was treated with intravenous hydralazine. Magnetic resonance imaging of his brain showed bilateral high T2 signal return in the parietal and occipital lobes in keeping with a radiological diagnosis of PRES. Complete resolution of the patient’s symptoms followed management of his emergency hypertension with intravenous hydralazine and maintenance with an ACE inhibitor. To our knowledge, this is the first reported case of PRES associated with low-renin, low-aldosterone hypertension.
A retrospective study of first permanent vascular access outcomes in incident haemodialysis patients
1University of Aberdeen, Aberdeen, UK
2Raigmore Renal Unit, Inverness, UK
Background
The successful provision of vascular access is fundamental for the effective delivery of haemodialysis (HD) in patients with established renal failure. However, vascular accesses are susceptible to the development of a range of complications. There is an immediate need to develop a greater understanding of the factors that influence access-related outcomes.
Aims
To determine the factors affecting first permanent vascular access outcomes in patients starting HD.
Methods
Data were retrospectively collected from the hospital’s surgical, radiological and renal database systems on factors related to patient characteristics and vascular access care for patients starting haemodialysis between 01 January 2007 and 31 December 2012 at Raigmore Hospital. The primary outcomes were primary and secondary survival of the first surgically created permanent access. Univariate analysis of access survival was undertaken for selected variables using the Kaplan–Meier method with log-rank testing used to test for differences between subgroups.
Results
Of the 128 patients starting haemodialysis at Raigmore renal unit between 01 January 2007 and 31 December 2012, 107 met the selection criteria and were included in the study; 25.2% (27/107) of included patients had previously failed peritoneal dialysis and 74.8% (80/107) were starting RRT for the first time with haemodialysis as their first treatment modality. In 96.3% (103/107) of patients, a permanent access was surgically created, of which 26.2%, 54.4% and 10.7% were radiocephalic, brachiocephalic and transposed basilic fistulae, respectively, and 8.7% were arteriovenous grafts. The reported complication and intervention rates for first permanent accesses were 1.3 and 0.8 per patient year on haemodialysis, respectively. Median primary survival of first access was 21.2 months (95% CI 14.3–28.1). Primary and secondary access survival rates at two years were 45.9% and 73.4%, respectively. A primary renal diagnosis of diabetes (p = 0.022), previous use of a temporary catheter (p = 0.003), rope-ladder needling (p = 0.013) and transposed basilic fistula or arteriovenous graft (p < 0.001) were predictive of significantly poorer permanent access survival.
Conclusion
The use of buttonhole needling may improve access survival.