Indian Journal of Cancer
Home  ICS  Feedback Subscribe Top cited articles Login 
Users Online :1343
Small font sizeDefault font sizeIncrease font size
Navigate here
Resource links
 »  Similar in PUBMED
 »  Search Pubmed for
 »  Search in Google Scholar for
 »Related articles
 »  Article in PDF (400 KB)
 »  Citation Manager
 »  Access Statistics
 »  Reader Comments
 »  Email Alert *
 »  Add to My List *
* Registration required (free)  

  In this article
 »  Abstract
 » Introduction
 »  Materials and Me...
 » Results
 » Discussion
 » Conclusion
 »  References
 »  Article Figures
 »  Article Tables

 Article Access Statistics
    PDF Downloaded230    
    Comments [Add]    

Recommend this journal


  Table of Contents  
Year : 2015  |  Volume : 52  |  Issue : 3  |  Page : 300-303

Do traumatic lumbar punctures lead to greater relapses in acute lymphoblastic leukemia? Experience at a university hospital in India

1 Department of Pediatrics, Hematology/Oncology Unit, Advanced Pediatric Center, Chandigarh, India
2 Department of Cytology and Gynecological Pathology, Post Graduate Institute of Medical Education and Research, Chandigarh, India

Date of Web Publication18-Feb-2016

Correspondence Address:
A Trehan
Department of Pediatrics, Hematology/Oncology Unit, Advanced Pediatric Center, Chandigarh
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/0019-509X.176722

Rights and Permissions

 » Abstract 

Objective: The aim of the study was to evaluate the impact of traumatic lumbar puncture (TLP) at diagnosis of relapse in childhood acute lymphoblastic leukemia (ALL). Risk factors associated with TLP were assessed. MATERIALS AND METHODS: A retrospective analysis was performed from the records of children with ALL who were treated from January 2010 to December 2012. Results: A total of 311 patients with median age of 5 years (range: 1–13) were treated for ALL. The cerebrospinal fluid analysis obtained from first LP revealed 275: Central nervous system 1 (CNS 1) (no blasts); 8: CNS 3 (blasts positive); and 28: TLP. Twenty-eight (9%) patients relapsed. Twelve (3.9%) had a CNS relapse. A TLP at diagnosis was not associated with an increased risk of systemic or CNS relapse (P = 0.298, 0.295). Three years event-free survival of patients with TLP and without atraumatic LP (ATLP) at diagnosis was 56 ± 5.2% and 51.8 ± 12.4%, (P = 0.520). Three years overall survival with TLP and ATLP was 73.3 ± 3.5% and 70.4 ± 12.5%, respectively, (P = 0.963). Median platelet count in patients with TLP was significantly lower than those without TLP (10,000/μL and 28,000/μL, P < 0.001). A receiver operating characteristic curve was constructed for predicting the risk of TLP based on platelet count. Area under the curve was 0.74 ± 0.05 (95% confidence interval 0.64–0.84). Platelet count < 23.5 × 109/L at the time of LP had 75% sensitivity and 64.4% specificity in predicting a TLP. Conclusions: Low platelet counts are significantly associated with risk of TLP. Traumatic LP at diagnosis was not associated with an increased risk of relapse.

Keywords: Acute leukemia, developing country, pediatric, relapse

How to cite this article:
Totadri S, Trehan A, Srinivasan R, Bansal D, Bhatia P. Do traumatic lumbar punctures lead to greater relapses in acute lymphoblastic leukemia? Experience at a university hospital in India. Indian J Cancer 2015;52:300-3

How to cite this URL:
Totadri S, Trehan A, Srinivasan R, Bansal D, Bhatia P. Do traumatic lumbar punctures lead to greater relapses in acute lymphoblastic leukemia? Experience at a university hospital in India. Indian J Cancer [serial online] 2015 [cited 2019 Sep 22];52:300-3. Available from:

 » Introduction Top

Central nervous system (CNS) directed therapy is a fundamental component in the management of pediatric acute lymphoblastic leukemia (ALL). It comprises intrathecal (IT) therapy and cranial irradiation.[1] In this era of diminishing use of cranial irradiation, optimum IT therapy is essential to prevent CNS relapse. Traumatic lumbar puncture (TLP), especially at the start of treatment, has been shown to translate into decreased event-free survival (EFS) and increased risk of CNS relapse.[2],[3],[4],[5],[6] The proposed mechanisms include seeding of the CNS “sanctuary” by peripheral blasts, potential masking of a CNS 3 status at diagnosis, and hematoma formation leading to compromise of subsequent IT therapy.[1] Thrombocytopenia is one of the significant risk factors contributing to TLP.[1],[7] This study was performed to determine the critical platelet count which would predict a TLP and the effect of TLP on outcome in pediatric ALL.

 » Materials and Methods Top

Patients treated for ALL from January 2010 to December 2012 were analyzed. The study was performed as a retrospective case record form review. Treatment was based on modified UKALL 2003 protocol. Diagnostic LP was performed in all patients. IT methotrexate (MTX) was administered on 2 fixed days of the week (Friday and Saturday) by trained pediatric residents. Procedural sedation with midazolam and ketamine was administered to all patients. Platelet transfusions were not administered to thrombocytopenic patients prior to LP or IT therapy unless they had mucosal/visceral bleeding or fever.

Cytomorphology was performed on the cerebrospinal fluid (CSF) samples obtained during each IT administration. CNS status was classified based on CSF cytomorphology as follows; (1) CNS 1: No blasts, (2) CNS 2: The presence of blasts with leukocyte count <5/µL, (3) CNS 3: The presence of blasts with leukocyte count ≥5/µL, (4) TLP: The presence of >10 red blood cell/µL, (5) CNS relapse: The presence of blasts and ≥5 leukocytes/µL in two successive CSF samples.

For the purpose of survival analysis following definitions was used; (1) EFS: Time from diagnosis to relapse or death, (2) overall survival (OS): Time from diagnosis to the time of last contact.

During the time period of the study, we had limited access to cytogenetics and minimal residual disease analysis. Patients were stratified as per the National Cancer Institute (NCI) Criteria into standard risk (SR) and high-risk (HR) patients based on the age and counts at admission (SR: Age 1–10 years and total leukocyte count [TLC] at presentation ≤50,000/µL, HR: Age > 10 years and/or TLC > 50,000/µL). T cell disease and Philadelphia (Ph) chromosome positivity were treated as HR with imatinib added for Ph-positive patients. The patients were treated as per the UKALL MRC 2001–2003 protocol for ALL. Patients received the standard and intermediate arm of the protocol for NCI standard and HR disease. This comprised three-drug induction for SR and four-drug induction for HR. Induction was followed by consolidation, interim maintenance, delayed intensification, and maintenance phase. Total therapy duration was 166 weeks in boys and 114 weeks in girls.

CNS-directed therapy included six doses of IT MTX during induction and consolidation, and three doses during delayed intensification. IT MTX was further administered every 3 months during maintenance. SR children (boys) received a total of 19 IT doses, and HR patients received 21 doses. Girls received three IT doses lesser than boys. Patients with a TLP at diagnosis received weekly IT during induction (2 extra IT as compared to CNS 1 patients).

Statistical methods

The data were entered in SPSS software (version 20.0, 2011, SPSS, Inc., Chicago, IL, USA). Baseline variables were analyzed by descriptive statistics. Comparison of proportions was done with Chi-square test and Fisher's exact test. Comparison of means between two different groups was performed by Student's t-test and Mann–Whitney U-test, for variables with normal and skewed distribution. The area under the receiver operating characteristic curve (AUC) was calculated for prediction of TLP by low platelet counts. Kaplan–Meier method was utilized for survival analysis.

 » Results Top

Three hundred and eleven patients were treated during the study period. Median age was 5 years (range: 1–13). Male: female ratio was 2.5:1. There were 190 (61%) SR children and 121 (39%) were treated as HR. CNS status at diagnosis was CNS 1: 275 (88.4%), CNS 3: 8 (2.6%), and TLP: 28 (9%) patients. Twenty-eight patients relapsed in total, among whom nine had isolated CNS relapse, and 3 had combined CNS and bone marrow relapse. CNS relapse occurred in 9, 2, and 1 patients in CNS 1, CNS 3, and TLP (P = 0.262). The mean follow-up duration of the cohort was 76.8 ± 47.7 weeks (range: 0.7–178).

Platelet count at the time of diagnostic LP was ≤100,000/µL in 264 (85.2%) patients. Platelet count was <30,000/μL and <20,000/μL in 180 (57.9%) and 138 (44.3%) patients. Clinical, investigational parameters and relapses of patients with and without TLP at diagnosis are compared in [Table 1]. Median platelet count in patients with TLP and atraumatic LP (ATLP) was 10,000/µL and 28,000/µL (P < 0.001). The receiver operating characteristic (ROC) curve was constructed using the total platelet count as the predictor variable and TLP as the dichotomous outcome. The ROC curve for low platelet count as a predictor of TLP had an AUC of 0.74 (95% confidence interval: 0.64–0.84), [Figure 1]. A platelet count of 23,500/µL at the time of LP had a sensitivity of 75% and specificity of 64.4% in predicting a TLP. In our cohort, platelet count was <23,500/µL in 150 (48.2%) patients. TLP was observed in 25 (8.6%) of 256 patients aged <10 years, and 6 (10.9%) of 55 aged ≥10 years (P = 0.604). White cell count at presentation and gender did not affect the occurrence of TLP [Table 1].
Table 1: Comparison of parameters between patients with TLP and ATLP at diagnosis

Click here to view
Figure 1: Receiver operating characteristic curve for studying the sensitivity and specificity of thrombocytopenia in predicting the occurrence of a traumatic lumbar puncture

Click here to view

A total of 3823 IT-s were performed in the 311 patients. There were 383 (10%) traumatic LPs. 128 patients did not have a single TLP during treatment with 14/128 having a relapse. In the 183 patients who had at least one TLP during treatment 14 relapsed (P = 0.323). Patients who had TLP at diagnosis went on to have a greater median number of TLP's during treatment (2.5, range: 0–7), as compared to patients with CNS 1 status at diagnosis (1, range: 0–7; P < 0.001).

Three-year EFS was 56% ±5.2% and 51.8% ±12.4% in patients with ATLP (CNS 1) and TLP at diagnosis (P = 0.520, [Figure 2]). Three-year OS was 73.3 ± 3.5% and 70.4 ± 12.5% in patients with ATLP and TLP at diagnosis (P = 0.963, [Figure 2]). Twenty-four out of 283 patients who had ATLP and 4 out of 28 patients who had TLP at diagnosis relapsed at any site (P = 0.298). Ten out of 283 patients who had ATLP, and 2 out of 28 patients who had TLP at diagnosis had an isolated CNS relapse (P = 0.295).
Figure 2: (a) Overall survival of patients who had a traumatic lumbar puncture and atraumatic lumbar puncture at diagnosis. (b) Event-free survival of patients who had traumatic lumbar puncture and atraumatic lumbar puncture at diagnosis

Click here to view

 » Discussion Top

LP with IT administration of chemotherapy is one of the most common procedures performed on children with ALL. TLP is not an uncommon problem and has been shown to correlate with increased relapse and inferior EFS.[2],[3],[4],[5],[6] The incidence of TLP is variable among different centers and ranges from 3% to 38%.[2],[3],[4],[5],[6],[7],[8],[9] The incidence of 9% in our study is within the reported range. Several pathophysiological mechanisms have been put forward to explain the contribution of TLP to increased relapse and reduced survival. In newly diagnosed patients, TLP can lead to the iatrogenic introduction of blasts from the peripheral blood into the CNS sanctuary site.[1],[5],[7] A TLP may imply that the needle was in an improper position prior to instillation of the IT chemotherapy leading to inadequate IT therapy.[5] In addition, the original CNS status at diagnosis can possibly be obscured by TLP resulting in potential under-treatment.[5],[7] A TLP could indirectly indicate advanced disease with perivenular and parameningeal leukemic infiltration, which may contribute to the reduced survival.[3],[6] The inadvertent pushing of the needle too far anteriorly can occur during a TLP, which may result in the penetration of the marrow cavity of the vertebral body. Subsequent repositioning and IT drug administration might flush adherent leukemic blasts from the marrow into the CSF.[3],[10]

Risk factors which are implicated in contributing to TLP include; age, race, thrombocytopenia, obesity, experience of the administering practitioner, hyperleukocytosis, and a previous TLP.[1],[4],[7],[8] Howard et al. and Shaikh et al. reported an increased risk of TLP in infants <1 year of age.[4],[7] Shaikh et al. observed a higher incidence of TLP among patients ≥10 years of age.[4] Cancela et al., on the other hand, found no significant association between age and TLP.[8] We had no infants in our analysis. Children older than 10 years did not have a higher incidence of TLPs. TLP did not differ between males and females, akin to other studies.[4],[7] Elevated white blood cell count was not associated with increased TLP in the studies performed by Shaikh et al. and Cancela et al. which was mirrored in our study.[4],[8]

Thrombocytopenia is an important risk factor which contributes to TLP. Howard et al. and Shaikh et al. reported an increased risk of TLP with platelet count <100,000/µL.[4],[7] In our study, patients with TLP had a significantly lower platelet count compared to those who had ATLP. Howard et al. have recommended transfusing platelets to raise the platelet count to ≥100,000/µL prior to diagnostic LP in patients with circulating blasts.[7] A cut-off of 100,000/µL would require platelet transfusions in a large number of patients diagnosed with ALL (85% of patients had a platelet count ≤100,000/µL in our study; with 44% and 58% having a count ≤20,000/µL and 30,000/µL). In the ROC curve constructed in our study, a platelet count of 23,500/µL offered a good trade-off for predicting TLP with 75% sensitivity and 64.4% specificity. This may be more feasible to ensure prior to diagnostic LP in resource-limited settings.

We observed that patients with TLP at diagnosis had a greater number of TLP's subsequently during treatment versus CNS 1 patients. A similar finding was reported by Howard et al., who further observed that shorter interval between a prior TLP and thrombocytopenia at the time of previous LP contributed to increased odds of TLP.[7]

Shaikh et al. (n = 264) reported 5 years EFS of 77% ± 8% and 93% ± 2% in patients with TLP with blasts in CSF at diagnosis (TLP+) and CNS 1 status (P = 0.002).[4] The Dutch Childhood Oncology Group (n = 526) demonstrated increased relapse at any site (P = 0.025) and poor 10 years EFS of 58 ± 7.6% in patients with TLP + status.[2] TLP without blasts (TLP−), on the other hand, was associated with superior EFS of 82 ± 5.2% (P < 0.01).[2] Gajjar et al. and Bürger et al. observed a similar inferior EFS in patients with TLP + as compared to CNS 1 and TLP−.[5],[6] In our study, there was no difference in EFS and OS between patients with TLP and those who did not have TLP at diagnosis. Rech et al. (n = 92) reported increased relapse only in HR-ALL with TLP.[3] In their study, TLP did not contribute to poor EFS or increased CNS relapse in the overall cohort.[3] The European Organization for Research and Treatment of Cancer Children Leukemia Group study 58881 (n = 2025) concluded that CNS status at diagnosis did not contribute significantly to prognosis.[9] Cancela et al. (n = 199) also had a similar observation that TLP did not contribute to increased cumulative incidence of relapse.[8] In our study, patients who had TLP at diagnosis received two extra IT doses of MTX during induction. This may have compensated for the deleterious effect of the TLP.

There were some limitations in our study. We did not differentiate between TLP+ and TLP−. The admixture of TLP− with TLP+ may have falsely improved the survival and decreased the relapse rate in the TLP cohort. We did not record details of the level of the clinician performing the procedure; there being evidence to suggest that greater experience decreases the odds of TLP.[7] In addition, the follow-up duration was short.

 » Conclusion Top

Thrombocytopenia is the only risk factor which significantly contributes to TLP. Patients who had TLP at diagnosis had a greater number of TLPs subsequently during therapy. In our study, TLP did not contribute to inferior EFS and increased relapse. Efforts should be made to ensure a platelet count >23,500/µL prior to doing an LP to avoid a TLP.

 » References Top

Pui CH, Howard SCs Current management and challenges of malignant disease in the CNS in paediatric leukaemia. Lancet Oncol 2008;9:257-68.  Back to cited text no. 1
Dutch Childhood Oncology Group, te Loo DM, Kamps WA, van der Does-van den Berg A, van Wering ER, de Graaf SS. Prognostic significance of blasts in the cerebrospinal fluid without pleiocytosis or a traumatic lumbar puncture in children with acute lymphoblastic leukemia: Experience of the Dutch Childhood Oncology Group. J Clin Oncol 2006;24:2332-6.  Back to cited text no. 2
Rech A, de Carvalho GP, Meneses CF, Hankins J, Howard S, Brunetto AL. The influence of traumatic lumbar puncture and timing of intrathecal therapy on outcome of pediatric acute lymphoblastic leukemia. Pediatr Hematol Oncol 2005;22:483-8.  Back to cited text no. 3
Shaikh F, Voicu L, Tole S, To T, Doria AS, Sung L, et al. The risk of traumatic lumbar punctures in children with acute lymphoblastic leukaemia. Eur J Cancer 2014;50:1482-9.  Back to cited text no. 4
Gajjar A, Harrison PL, Sandlund JT, Rivera GK, Ribeiro RC, Rubnitz JE, et al. Traumatic lumbar puncture at diagnosis adversely affects outcome in childhood acute lymphoblastic leukemia. Blood 2000;96:3381-4.  Back to cited text no. 5
Bürger B, Zimmermann M, Mann G, Kühl J, Löning L, Riehm H, et al. Diagnostic cerebrospinal fluid examination in children with acute lymphoblastic leukemia: Significance of low leukocyte counts with blasts or traumatic lumbar puncture. J Clin Oncol 2003;21:184-8.  Back to cited text no. 6
Howard SC, Gajjar AJ, Cheng C, Kritchevsky SB, Somes GW, Harrison PL, et al. Risk factors for traumatic and bloody lumbar puncture in children with acute lymphoblastic leukemia. JAMA 2002;288:2001-7.  Back to cited text no. 7
Cancela CS, Murao M, Viana MB, de Oliveira BM. Incidence and risk factors for central nervous system relapse in children and adolescents with acute lymphoblastic leukemia. Rev Bras Hematol Hemoter 2012;34:436-41.  Back to cited text no. 8
Sirvent N, Suciu S, Rialland X, Millot F, Benoit Y, Plantaz D, et al. Prognostic significance of the initial cerebro-spinal fluid (CSF) involvement of children with acute lymphoblastic leukaemia (ALL) treated without cranial irradiation: Results of European Organization for Research and Treatment of Cancer (EORTC) Children Leukemia Group study 58881. Eur J Cancer 2011;47:239-47.  Back to cited text no. 9
Kruskall MS, Carter SR, Ritz LP. Contamination of cerebrospinal fluid by vertebral bone-marrow cells during lumbar puncture. N Engl J Med 1983;308:697-700.  Back to cited text no. 10


  [Figure 1], [Figure 2]

  [Table 1]


Print this article  Email this article


  Site Map | What's new | Copyright and Disclaimer
  Online since 1st April '07
  © 2007 - Indian Journal of Cancer | Published by Wolters Kluwer - Medknow