Brochures

Renal Effects of Long-term Lithium Therapy

Description
Lítio, efeitos a longo prazo
Categories
Published
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Renal effects of long-term lithium therapy in the elderly:a cross-sectional study Els J. M. van Melick  1 *, Arend E. Meinders 2 , Tonko O. Hoffman 3 and Toine C. G. Egberts 4,5 1  Department of Geriatrics, Parnassia Psycho-Medical Centre, The Hague, The Netherlands 2  Department of General Internal Medicine, Leiden University Medical Centre, Leiden, The Netherlands 3  Department of Research, Parnassia Psycho-Medical Centre, The Hague, The Netherlands 4  Department of Pharmacoepidemiology and Pharmacotherapy, Utrecht Institute of Pharmaceutical Sciences, Faculty of Science, Utrecht University, Utrecht, The Netherlands 5  Department of Clinical Pharmacy, University Medical Centre Utrecht, Utrecht, The Netherlands SUMMARY Objectives  To determine the effect of long-term lithium therapy on glomerular filtration rate (GFR) and maximum renalconcentrating capacity (Umax) in the elderly, to identify possible risk factors, to determine the clinical impact of a reducedUmax in this population and in case of polyuria to establish a diagnosis. Methods  This is a cross-sectional study with 48 outpatients of 65 years or over (mean 74.8 years), who were treated withlithiumformorethan6months(mean9.2years).TheGFRwasdeterminedwiththeCockcroft-Gaultformula(GFR-CG)andthe Umax was measured in a urine sample collected between 3 and 5h after the patients received 40 m g desmopressin(DDAVP) intranasally. Results  No relation was found between duration of lithium treatment and GFR-CG, but there was a significant negativerelationbetween durationoflithiumtreatmentandUmax(B  0.73;CI:  1.249/   0.212);73%ofthepatientshadamoderatetosevereconcentratingdefect.Nootherriskfactorsthandurationoflithiumtherapywereidentified.AreducedUmaxcausedpolyuria ( > 2500mL/24h) in 33% but did not cause significant more thirst, incontinence or disturbed sleep. Conclusions  In this geriatric population a negative relation was found between duration of lithium treatment and Umax.But a reduced Umax did not result in significant more clinical symptoms. In case of polyuria other mechanisms besidenephrogenic diabetes insipidus were found to play a role in this age group. Copyright # 2008 John Wiley & Sons, Ltd. key words —lithium; elderly; glomerular filtration rate; renal concentrating capacity; arginine vasopressin INTRODUCTIONMore than 57 years after the discovery of thetherapeutic effects of lithium, it is still an importantdrug that is also used for people of 65 years and over(Cade, 1949). In a cross-sectional study Head andDening (1998) found a prevalence of lithium use inolderadultsof0.27%.Thisismorethantheprevalenceoflithium useinpeopleof allages and probablyduetothe necessity of long-term treatment. The mostimportant indications for lithium treatment in theelderly are bipolar disorder and unipolar depression(Bech, 2006). It is well known that lithium can cause adecline in maximum renal concentrating capacity(Umax) especially after long-term use (Vestegaard et al ., 1979; Boton  et al ., 1987; Waller and Edwards,1988). Risk factors for a more serious reduced Umaxare concomitant use of other pyschotropics or somaticmedication (Bendz  et al ., 1994).The effect of lithium on glomerular filtration rate(GFR) is less clear. In most older studies there isevidence for a mild decline at the most, though Bendz et al . (1994) found in their study an age-adjustedreduction in GFR of 21%. From recent work there is INTERNATIONAL JOURNAL OF GERIATRIC PSYCHIATRY  Int J Geriatr Psychiatry  2008;  23 : 685–692.Published online 10 January 2008 in Wiley InterScience(www.interscience.wiley.com)  DOI : 10.1002/gps.1961*Correspondence to: Dr E. J. M. van Melick, Parnassia Psycho-MedicalCentre,DepartmentofGeriatrics,Mangostr1,2552KSTheHague, The Netherlands. E-mail: e.van.melick@parnassia.nlCopyright # 2008 John Wiley & Sons, Ltd.  Received 27 June 2007  Accepted 31 October 2007   evidence that lithium can also cause end-stage renaldisease but the incidence is probable very low(Markowitz  et al ., 2000; Presne  et al ., 2003).Many studies have investigated the renal effects of lithium in groups of patients of all ages but, to ourknowledge, this has never been studied specifically inan elderly population. The aim of this study wasprimary to investigate the relationship between theduration of lithium use and Umax and GFR in peopleof 65 years or more and to identify possible risk factors. Secondly to evaluate if a reduced Umaxcauses clinical symptoms or impact on daily activitiesinelderlylithiumusers.Finally,toinvestigatethemostlikely diagnosis in these elderly patients in case of polyuria.METHODS Setting and study population The study had a cross-sectional design and wasconducted in the spring of 2005. The study protocolwas approved by the medical-ethical committee of Dutch psychiatric institutions. The study populationconsisted of patients treated in Parnassia, a psychiatricinstitution with hospital and ambulatory facilitieslocated in The Hague, The Netherlands, servingapproximate 40.000 patients a year.All ambulatory patients of 65 years and over whowere treated with lithium for more than six monthsreceived an explanatory letter about the study.Exclusion criteria were severe urinary incontinence(the patient can not control his/her voiding in any wayand therefore can not collect even a small sample of urine), inability to understand Dutch and cognitiveimpairment. Of the 186 patients listed as user of lithium, 48 participated in the study (Figure 1).  Measurements The 48 patients who participated in the study werevisited on study day one by a research nurse or adoctor. A questionnaire about polyuria-related pro-blems was filled out: questions about how many timesa patient had to void during the day and during thenight and if thevoiding frequency interfered with theirsocial functioning and/or sleep and questions aboutthirst and incontinence. A separate questionnaireabout lithium use, use of other medication and currentand past illnesses was also filled out. Thesequestionnaires were checked and if necessary com-pleted with information from the medical record. Allpatients collected 24h urine (V24) on day 2 from8.00h am until day 3 8.00h am. In the morning of day 3 the patients came to the hospital, where they wereweighted and blood samples were collected for themeasurement of serum lithium concentration 12h postdosing, urea nitrogen, creatinine, natrium, kalium,magnesium, calcium, phosphate, albumine, glucose,TSH, fT4, T3 and arginine vasopressin (AVP). Next40 m g desmopressin (DDAVP), a synthetic argininevasopressin derivate, was administered intranasally(Monson and Richards, 1978; Tryding  et al ., 1987).The 24-h creatinine excretion was used todetermine the completeness of the urine collection(Rose, 2001). A sample of the 24h urinewas taken forcreatinine, natrium, protein, glucose and culture.Another urine sample was collected for the determi-nation of the urine osmolality (Umax) at least 3h butno longer than 5h after DDAVP was administered.There was no fluid restriction during the study period(Monson and Richards, 1978; Tryding  et al ., 1988).Plasma AVP was determined using a radio-immunoassay (Jodium 125) after plasma extractionover a C 18  column. The AVP measurements had anintra-assay CV between 5.4 and 11.2% with concen-trations resp 4.15 and 0.49ng/l and the inter-assay CVwas between 2.8 and 4.4% with concentrations resp0.5 and 4.0ng/l.Urine osmolality was determined using an osm-ometer based on the principle of freezing pointdepression. The urine-osmolality measurements hadan intra- and interassay CV of 1%. Figure 1. Patient selection.Copyright # 2008 John Wiley & Sons, Ltd.  Int J Geriatr Psychiatry  2008;  23 : 685–692.DOI: 10.1002/gps 686  e. j. m. van melick  ET AL.  Plasma osmolality (Posmol) was calculated accord-ing to the following formula: 2[Na þ ] (mmol/ L) þ urea(mmol/L) þ glucose(mmol/L).TheGFRwasestimatedbyusingtheCockcroft-Gaultformula (GFR-CG) (Cockcroft and Gault, 1976;Lamb  et al ., 2003). As Umax declines with age thereference values according to Tryding  et al . (1988)were used. He stated that at the age of 80 thelowest maximum urine osmolality (mean–2SD) is600mosmol/kg. No reference values are, to ourknowledge, available for patients  > 80 years.The possible risk factors for a reduced Umax westudied were: vascular disease, vascular risk factors,the concomitant use of antispsychotics, use of different antidepressants, antihypertensiva and thetotal number of drugs currently used (Bendz  et al .,1994; Movig  et al ., 2003).To determine the clinical impact of a reduced Umaxwe looked for problems that could be related topolyuria. This was studied with the above mentionedquestionnaire. As a reduced Umax with polyuria canincrease the risk for dehydration and thereby forlithium intoxication this was also registered for everypatient (Oakley  et al ., 2001).To establish a diagnosis in patients with polyuria( > 2500mL/24h) the Posmol, Umax, plasma AVP andV24 were combined for diagnosing the individualpatient. If the patient did not collect a completeV24 according to the creatinine excretion we made acorrection (cV24) (Rose and Post, 2001). Expectedcreatinine excretion for a man [(28–(age/6) ¼ . . . mg/ kg/dag  bodyweight  8.840  10  3 . . . mmol/day]/ measured creatinine excretion mmol/L ¼ cV24. For awomen the formula starts with 22–(age/9).The patients were not subjected to a thirst test.Those with a normal Posmol and decreased Umax,could still have a significant clinical problem. WeassumedthatthesepatientsshouldhaveaV24orcV24of at least > 2500mL. There are no absolute criteria inthe literature available for the diagnosis of nephro-genic diabetes insipidus (NDI), central diabetesinsipidus (CDI) and primary polydipsia (PP). Thecriteria we applied are given in Table 1 and arepartially derived from Rose and Post (2001).  Data analysis To examine the relation between duration of lithiumuse, GFR-CG and Umax, two hierarchical multiplelinear regression analyses were used. In the firstregression GFR-CG was used as dependent variableand in the second regression Umax each with durationof lithium use as predictor. In both analyses, age wasalso included as predictor in the first block andvascular diseases and vascular risk factors wereincluded in the second block using the method‘ENTER’. Use of Calcium-antagonists, use of ACE-inhibitors and use of thiazide and loop diureticswereincludedaspredictorsinthethirdblock,butweredropped due to non-significance, except for use of Calcium-antagonists.Toexaminethe relationbetweenduration of lithiumuse and AVP concentration a hierarchical multiplelinear regression analyses was used with AVP asdependentvariableanddurationoflithiumuseandageas predictors.To determine the clinical impact of a reduced Umaxwe studied the problems related to polyuria for threegroups: severe reduction (  300mosmol/kg), mildto moderate reduction (300msomol/kg  < Umax  600msomol/kg) and no reduction of Umax( > 600mosmol/kg). A chi-squarewas used to examinethe distribution within the categories of Umax. Anexact correction for small expected values was usedwhen needed (Agresti, 2002). A logitmodel was notused due to small sample sizes. All calculations werecarried out using SPSS version 13.0. Table 1. Diagnostic criteria in case of polyuria 18 Diagnosis V24 o´r cV24* (mL) Umax** (mosmol/kg) Posmol (mosmol/kg) AVP (ng/L)NDI   3000   300  > 290  > 1.5pNDI  > 2500   600  > 285  > 1.5CDI   3000  < 400  > 290   0.5pCDI  > 2500  < 700  > 285   0.5PP  > 2500   600  < 285  < 0.5*V24 or cV24 24-h urine production measured or corrected according to creatinine excretion.**Umax maximum renal concentrating capacity after DDAVP.AVP ¼ Arginine Vasopressin; CDI ¼ Central Diabetes Insipidus; NDI ¼ Nephrogenic Diabetes Insipidus; pCDI ¼ partial Central DiabetesInsipidus; pNDI ¼ partial Nephrogenic Diabetes Insipidus; Posmol ¼ Plasma osmolality, PP ¼ Primary Polydipsia.Copyright # 2008 John Wiley & Sons, Ltd.  Int J Geriatr Psychiatry  2008;  23 : 685–692.DOI: 10.1002/gps renal effects of long-term lithium therapy in the elderly  687  RESULTSIn this study 48 patients participated with a mean ageof 74.8 years (Table 2).Depression was the most frequent reason forlithiumuse (62.5%), followed by bipolar disorder(35%). This can be explained by the fact that lithiumaugmentation for treatment of resistant unipolardepression was a long-standing practice in Parnassialong before this was adviced in a national guideline(Multidisciplinaire richtlijn depressie, 2005).All patients were on lithium for more than 6 monthswith a mean of 9.2 years. They took their lithiummedication once a day in the evening. The advicedlithium level for the elderly was 0.4–0.6mmol/L forstable patients and up to 0.8mmol/L in acute patients.Polypharmacy was a frequent finding with 52% of patients using five or more medications a day. This ishigher than in the same age group in the generalpopulation (Heerdink, 1995).We found no significant relation between theduration of lithium treatment and the GFR-CG butthere was a significant negative relation betweenduration of lithium use and Umax (Table 3). TheUmax was reduced in 73% of the patients; 19% had asevere concentrating defect (  300mosmol/kg) and54% had a moderate concentrating defect ( > 300 and  600mosmol/kg). None of the patients had abnormalplasma concentrations of Ca 2 þ or K  þ and none hadglucosuria. One patient had proteinuria of 0.3g/L, theothers had lower or no proteinuria at all.In this patient group we found no significantinfluence of vascular disease, vascular risk factors,concomitant use of SSRIs, TCAs, antipsychotics orantihypertensives or of the total number of medi-cations on the Umax in relation to duration of lithium Table 2. Demographic and medical characteristics of the study populationCharacteristics CharacteristicssubdivisonMean (range) SD AbsolutenumberPercentageGender Men 7 15Women 41 85Age (years) 74.8 (65–89) 5.65Bodyweight (kg) 76.6 (41–121) 14.37Duration lithium use (years) 9.2 (0.5–31) 8.58Serum lithium concentration 12h post dosing (mmol/L) 0.61 (0.13–0.86) 0.138Daily lithium dose (mg lithiumsalt) 470.65 (150–800) 149.66Indication lithium use Bipolar disorder 17 35Depression 30 63Other 1 2Vascular disease and vascular risk factors 30 63Cardiovascular disease 6 13Cerebrovascular disease 5 10Diabetes Mellitus 7 15Hypertension 21 44Hyperlipidaemia 5 10Known renal disease 2 4Hypothyroı¨dism* 17 35Total number medication 4.8 (1–9) 2.27Antidepressants 29 60TCA 19 40SSRI 7 15Other 4 8Antipsychotics 6 13Atypical 5 10Classic 1 2Loopdiuretics 3 6Thiazide type diuretics One with amiloride 5 10NSAIDs 0 0ACE-inhibitors 7 15Calcium-antagonists 4 8*Hypothyroidism defined as patients who were treated with thyroxine, all other patients had normal fT4 and TSH, except two who hadsubclinical hypothyroidism with normal fT4 and TSH > 4.0 and < 10.0mU/L.ACE-inhibitor ¼ Angiotensin Converting Enzym-inhibitor; NSAID ¼ Non-Steroidal Anti-Inflammatory Drug; SSRI ¼ Selective SerotoninReuptake Inhibitor; TCA ¼ Tricyclic Antidepressant.Copyright # 2008 John Wiley & Sons, Ltd.  Int J Geriatr Psychiatry  2008;  23 : 685–692.DOI: 10.1002/gps 688  e. j. m. van melick  ET AL.
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks