Sickle cell disease-related organ damage occurs irrespective of pain rate: implications for clinical practice

In daily clinical practice, the frequency of painful crises(pain rate) is an important parameter of sickle cell diseaseseverity. We assessed the prevalence of sickle cell disease-relatedorgan damage and complications and their relation to pain rate.Organ damage and history of vaso-occlusive complications wereobtained via systematic screening of consecutive patients andby chart review. In 104 adult sickle cell patients pain ratewas related to a history of acute chest syndromes, avascularosteonecrosis, iron overload, priapism and cholelithiasis. However,major disease-related complications, such as microalbuminuriaand pulmonary hypertension, were detected in 23% and 24% respectivelyof patients without painful crises in the study period underliningthe importance of systematic screening for developing organdamage in sickle cell patients irrespective of pain rate.

Key words: sickle cell disease, organ damage, pain rate, systematic screening.

Introduction

TOP

Introduction

Sickle cell disease (SCD) is heterogeneous in its presentation,with differences in the rate and severity of complications evenwithin a single genotype. Even patients with the most severegenotype, HbSS, may vary in their clinical presentation frombeing continuously admitted for the management of acute complicationsto rarely requiring medical care. Both vaso-occlusion and chronichemolysis are major determinants of SCD-related organ damage.¹With the increasing life expectancy of sickle cell patientsin the Western world the effect of accumulating organ damageon the quality of life and life expectancy is becoming an importantfactor in managing SCD.² Early recognition of developing organdamage is imperative in order to institute specific therapeuticsin a timely manner. However, a landmark autopsy study in sicklecell patients demonstrated a high prevalence of organ damagethat often had not been recognized during life by treating phycisians.²Although the frequency of the painful sickle cell crisis, whichis the hallmark SCD-related clinical complication, is considereda parameter of disease severity, most patients do not frequentlyexperience painful crises that require medical care. Nonetheless,sickle cell patients generally have a significantly reducedlife expectancy suggesting that clinically significant organdamage accumulates irrespective of the pain rate.⁴ As the painrate is considered an important parameter of SCD severity, weanalyzed whether the prevalence of SCD-related manifestationsis related to the frequency of painful crises.

Design and Methods

TOP

Design and Methods

Patients
Adult sickle cell patients visiting the Department of Haematologyof the Academic Medical Center (AMC) in Amsterdam were consideredeligible. After obtaining written and informed consent, patientswere screened for SCD-related manifestations from July 2005until December 2006. This study was approved by the internalreview board of the AMC and carried out in accordance with theprinciples of the Declaration of Helsinki.

SCD related manifestations
SCD-related manifestations were assessed by systematic screeningand medical record review and defined as follows: microalbuminuria:urinary creatinine (mmol/L) to urinary albumin (mg/L) ratio>3.5 (males)/>2.5 (females) confirmed with 24 hours urinecollection with microalbuminuria >30 mg/24 hours.⁵ Renalfailure: creatinine clearance <100 mL/min (Cockcroft andGault).^6,7 Pulmonary hypertension (PHT): tricuspid regurgitationjet flow velocity (TRV) $≥$ 2.5 m/sec in rest detected by Dopplerechocardiography. PHT was considered absent with no or onlytrace TRV.⁸ Retinopathy: presence of at least mild non-proliferativeretinopathy.⁹ Perceptive hearing loss: loss of >20 dB withno other explanation than SCD.¹⁰ Cholelithiasis: presence ofgallstones (ultrasound) or previous cholecystectomy becauseof chole-cystolithiasis. Iron overload: plasma ferritin level>1000 µmol/L (on at least three occasions during steadystate) and a history of >20 transfused packed cells.¹¹ Acutechest syndrome: defined as previously described¹² occurringbetween January 2002-January 2007. Symptomatic avascular osteonecrosis:local pain and reduced function with documented osteonecrosisof the femoral or humeral head (hip or shoulder X-ray) or ahistory of surgical intervention for osteonecrosis. Leg ulcers:chronic ulcers of the ankle not otherwise explained. Priapism:spontaneous painful erection requiring hospital care. Stroke:history of stroke confirmed by magnetic resonance imaging orcomputerized tomography.

Pain rate
Pain rate was assessed by calculating the cumulative numberof admissions for painful crises (defined as typical musculoskeletal/abdominalpain not otherwise explained)⁴ from January 2002 until January2007 and categorizing patients into 3 groups: 0 crises/year,>0 and <1 crises/year or $≥$ 1 crises/year. Painful crisesnot requiring medical care were excluded.

Laboratory parameters
All laboratory data were obtained during routine out-patientvisits at least four weeks after the last acute disease-relatedcomplication or blood transfusion. Fetal hemoglobin percentage(HbF%) was determined by cation-exchange high performance liquidchromatography,¹³ and ${alpha}$ -thalassemia screening was performed witha multiplex PCR assay.¹⁴

Statistical analysis
The most severe SCD genotypes (HbSS and HbSβ⁺-thalassemia)were grouped together, as were the relatively severe genotypes(HbSC and HbSβ⁺-thalassemia). Continuous data are presentedas medians with their corresponding interquartile range. Betweengroup differences were tested with the Mann-Whitney U test.Categorical data are presented as percentages with between groupdifferences or statistical dependence tested with Fishers’Exact Test. Bivariate correlations of ordinal data were testedby determining the Spearman correlation coefficient (rs). pvalues below 0.05 were considered statistically significant.SPSS 12.0.2 (SPSS Inc, Chicago, IL, USA) was used.

Results and Discussion

TOP

Results and Discussion

One hundred and ten adult sickle cell patients were eligibleof whom 6 were excluded due to incomplete data collection (Table 1).Apart from retinopathy (which was more prevalent in HbSC/HbSβ⁺-thalassemiapatients), most manifestations of SCD were significantly moreoften present in HbSS/HbSβ⁰-thalassemia patients. PHT wasdetected in 32% and 12% of the HbSS-HbSβ⁰ and HbSC/HbSβ⁺respectively, with a median TRV of 2.60 (2.50–2.69) m/sec.None of these patients had severe PHT (TRV>3.0 m/sec.). Althoughsignificantly more patients with frequent sickle cell crisesused hydroxyurea, no difference in SCD-related organ damagewas observed between patients with or without hydroxyurea (datanot shown).

View this table:
[in this window]
[in a new window]
[Download PPT slide]
Table 1. Patients’ characteristics.

Avascular osteonecrosis, a history of acute chest syndrome,priapism and cholelithiasis, as well as iron overload were significantlyrelated to pain rate (Table 2). The association between ironoverload and the pain rate is probably the result of liberalblood transfusions for treating painful crises prior to institutingevidence-based management protocols for SCD in the Netherlands.¹⁵Importantly, microalbuminuria and PHT were detected in 23% and24% respectively of patients without painful crises during thestudy period. Furthermore, PHT and microalbuminuria were detectedin 23% and 10% respectively of patients with no painful crisesin the last five years. These patients had no leg ulcers, episodesof priapism or acute chest syndrome in the last five years andappeared clinically well according to case history and physicalexamination. Such patients would probably have been misclassifiedas having mild SCD. These data indicate that several major diseaserelated complications^6,10 are not related to pain rate and occureven in a significant number of patients that seem clinicallywell. This underlines the importance of systematic screeningfor SCD-related complications even in clinically mildly affectedpatients.

View this table:
[in this window]
[in a new window]
[Download PPT slide]
Table 2. Prevalence of sickle cell related complications.

Several shortcomings of this study need to be addressed. Firstly,the history of acute painful crises was limited to the lastfive years and only painful crises for which patients were admittedwere evaluated. Therefore, the conclusions may not be extrapolatedfor the number of painful crises experienced at home or beforethe evaluated five year period. Secondly, the retrospectivenature of this study has probably resulted in a selection bias.Thirdly, since this study was performed in a tertiary teachinghospital, referral bias cannot be excluded. However, given thesimilar prevalence of most SCD-related disease manifestationsin our cohort to that reported in literature^4,8,16–23the study seems to be representative. The prevalence of renalfailure may, however, be underestimated as supranormal proximaltubular function characteristic of SCD probably results in anoverestimation of glomerular filtration.²⁴ Also, the prevalenceof retinopathy in our study is higher than that of previousreports which is probably due to the inclusion of mild non-proliferativeretinopathy.²⁵ Lastly, other forms of sickle cell-related organdamage, such as pulmonary, hepatic and neurocognitive organdamage, have not been analyzed in this study. Nonetheless, wefeel that the aforementioned factors do not influence the mainfindings of our study.

In conclusion, clinically relevant forms of organ damage, suchas PHT and micro-albuminuria, occur irrespective of the frequencyof painful crises in adults with SCD. Systematic screening forand evaluation of organ damage in all sickle cell patients seemsindicated since many of the sickle cell-related complicationsmay otherwise go unnoticed, thereby delaying the institutionof potential therapeutic measures.

Footnotes

The CURAMA study group is a collaborative effort studying sicklecell disease in the Netherlands Antilles and the Netherlands.Participating centers: The Red Cross Blood Bank Foundation,Curaçao, Netherlands Antilles; The Antillean Institutefor Health Research, Curaçao, Netherlands Antilles, TheDepartment of Internal Medicine, Slotervaart Hospital, Amsterdam,the Netherlands; the Department of Vascular Medicine and theDepartment of Hematology, Academic Medical Center, Amsterdam,the Netherlands; the Department of Hematology, Erasmus MedicalCenter, Rotterdam, the Netherlands; the Department of Pathology,Groningen University Hospital, the Netherlands; the Departmentof Internal Medicine, the Laboratory of Clinical Thrombosisand Hemostasis, and the Cardiovascular Research Institute, AcademicHospital Maastricht, the Netherlands.

Authorship and Disclosures

EJvB, JBS and BJB wrote the article, AvdG was involved in gatheringpatient data and MRMG gave methodological and statistical adviceand was specifically involved in the echocardiographic evaluationof the patients. The authors reported no potential confictsof interest.

Received for publication August 23, 2007. Revision received November 17, 2007. Accepted for publication December 5, 2007.

References

TOP