• stem cell transplantation;
  • late adverse effects;
  • permanent teeth;
  • tooth development;
  • dental root development;
  • root-crown ratio;
  • children;
  • chemotherapy;
  • irradiation


  1. Top of page
  2. Abstract
  6. Acknowledgements


Deficient dental root development has been reported after conventional pediatric anticancer therapy, but less information is available on stem cell transplantation (SCT) recipients.


Root-crown (R/C) ratios of fully developed permanent teeth were assessed from panoramic radiographs of 52 SCT recipients, who were treated when they were age < 10 years. Using standard deviation scores (SDSs), the authors compared the R/C ratios to the corresponding tooth and gender-specific values in a healthy population. The percentage of affected R/C ratios per individual was examined in a subgroup of 39 (SG39) patients with advanced tooth development. The effects of total body irradiation (TBI) and SCT age on the R/C ratios were studied in TBI and high-dose chemotherapy (HDC = non-TBI) groups and in 3 age groups (≤ 3.0 years, 3.1–5.0 years, ≥ 5.1 years).


Per individual, 77% of the fully developed permanent teeth were affected in SG39. At the tooth level, in 77% of the 945 teeth studied (52 patients), the R/C ratios were outside ±2 SDSs. More teeth were affected in the TBI (85%) than in the non-TBI (55%) group (P < 0.001). The teeth of the patients who were ages 3.1–5.0 years old at SCT presented with the most severe aberrations of the R/C ratio (mean SDS = −4.4) whereas the teeth of the youngest (age ≤ 3.0 years) and the oldest (age ≥ 5.1 years) patients were equally affected (mean SDSs = −3.1 and −3.0, respectively).


Disturbances of dental root growth always followed pediatric SCT. HDC alone intensely harmed root growth but TBI further increased the adverse effects that were most extensive in the patients 3.1–5.0 years at SCT. These sequelae should be taken into account during the lifelong dental follow-up to minimize the clinical consequences of dental injuries. Cancer 2005. © 2005 American Cancer Society.

Late effects of treatment for childhood cancer may involve cardiovascular,1, 2 endocrine,3, 4 metabolic,3 gastrointestinal,5, 6 and musculoskeletal7, 8 systems. Growth disturbances are also common.9–11 Several kinds of aberrations in dental development have been described, including tooth agenesis, microdontia, and disturbed development of dental roots.12–23 Mostly study populations have consisted of children with cancer, treated with conventional chemotherapy, sometimes combined with central nervous system (CNS) irradiation in patients with leukemia and focal irradiation and/or surgery in patients with solid tumors. To our knowledge, very few studies to date have investigated dental consequences in recipients of pediatric stem cell transplantations (SCT).15, 17, 19, 24, 25 Intensive chemotherapy and radiotherapy preceding SCT and young age at SCT (< 5 or 6 years) have been observed in clinical studies to result in the worst dental consequences. The agenesis prevalence of permanent teeth has been 31–80%, microdontia prevalence 44–78%, and aberrations in dental root development have been very common.15, 17, 19, 25 Preparative regimens for SCT, i.e., total body irradiation (TBI) and high-dose chemotherapy (HDC), are considered to be responsible for the developmental dental disturbances.

The adverse effects of chemotherapy and/or radiotherapy on dental root development include failure of root development, short roots, V-shaped tapering roots (Fig. 1B,C), and blunting of the apical area. Alterations of root development mostly have been listed by a descriptive, still subjective method,16, 17, 19 originally used by Dahllöf et al.,15 or, in some studies, the methods have not been described.20, 21, 24, 26, 27 Studies using objective measurements to quantify the developmental aberrations are rare. Area measurements of dental crowns and/or roots, ratios of root and crown areas,14, 23, 28 or root-crown (R/C) ratios based on tooth length measurements25 have been used. Calcification of permanent teeth begins at the time or soon after the birth, taking approximately 15 years when the third molars are excluded.29, 30 After the completion of dental crown development, the cells of the Hertwig epithelial root sheath initiate dental root development.31, 32 The first signs of root development in permanent teeth are seen on panoramic radiographs (PRGs) starting from the age of approximately 3 years (central incisors, permanent first molars) to 7.5 years of age (permanent second molars).33 Environmental insults may alter the normal pattern. Because tooth development is a slow process, a latency period of 1–2 years can be expected before the first signs of possible disturbances are detectable on radiographs. Depending on the (dental) age and the treatment preceding SCT, the adverse effects may be directed differentially on different teeth. Thus, the same treatment may lead to dental agenesis or microdontia at an early stage of tooth development but, later on, disturb root development. The ultimate effect of anticancer therapy is not seen until the growth of dental roots has been completed, often several years after the treatment. In addition to age, the cytotoxic agents and their dosages and radiotherapy are of importance, as their varying ability to cause sublethal or lethal damage to tooth-forming cells contributes to the clinical outcome of the aberration. Unlike other bony structures, teeth do not remodel, and the developmental dental disturbances are permanent.

thumbnail image

Figure 1. (A) Permanent left mandibular teeth of a healthy male (third molars excluded). All root-crown (R/C) ratios are in the normal area of ±2 standard deviation scores (SDSs) when compared with the mean values previously reported in males.47 (B) Permanent left mandibular teeth of a female who was treated for acute lymphoblastic leukemia and received an allogenic stem cell transplantation (SCT) at the age of 5.0 years. Fractionated total body irradiation (TBI) of 10 grays (Gy) was included in the preparative regimen. Compared with the mean R/C ratios of healthy females, the SDSs from the central incisor to the second molar are −6.0, −5.9, −2.2, −3.7, −2.6, −2.4, and −3.6, respectively. (C) Permanent left mandibular teeth of a male who was treated for neuroblastoma and received an autologic SCT at the age of 4.1 years. Fractionated TBI of 12 Gy was included in the preparative regimen. Compared with the mean R/C ratios of healthy males, the SDSs from the central incisor to the second molar are −5.1, −6.2, −5.0, −3.4 (P: second premolar is missing and primary molar persists), −8.7, and −7.0, respectively.

Download figure to PowerPoint

Histologic studies on extracted teeth after chemotherapy or irradiation in childhood have revealed accentuated incremental lines in dentin26, 34, 35 and sometimes in enamel,35 synchronous with the anticancer treatment cycles. The disturbing effect of several chemotherapeutic agents and irradiation on developing teeth also has been revealed in animal and in vitro studies.36–46

In the current study, we assessed the R/C ratios of all fully developed permanent teeth in SCT recipients. Using standard deviation scores (SDSs), we compared them to the gender-specific “normal” R/C values of the corresponding teeth obtained from a healthy white (Finnish) population by the same assessment method.47 To study the effects of age and TBI on the R/C ratios, patients and teeth were divided into subgroups based on TBI and age at SCT.


  1. Top of page
  2. Abstract
  6. Acknowledgements


We examined the root development of permanent teeth from PRGs in the group of SCT recipients, who had been transplanted when they were age < 10 years. The patients had been treated at the Hospital for Children and Adolescents, University of Helsinki (Helsinki, Finland) between 1980 and 1999. Sixty-six percent (n = 56) of the eligible survivors volunteered to take part in the examination. Four patients were later excluded, as they had no definable teeth due to a young age, and the final study group comprised 52 patients (26 males and 26 females). The mean age of the patients was 4.4 years (range, 1.0–9.4 years) years at SCT and 11.7 years (range, 4.7–25.7 years) at the examination. The mean follow-up time was 7.2 years (range, 1.0–20.6 years) after SCT. The diagnoses of the patients are listed in Table 1. Twelve patients, seven with acute lymphoblastic leukemia (ALL), two with acute myeloblastic leukemia (AML), and one each with lymphoma, Wilms tumor, and rhabdomyosarcoma (RMS) received transplants due to a recurrent disease. The treatment periods from diagnosis to SCT varied from 0.2 to 2.6 years (mean, 0.8 years). Except for the diseases mentioned above, none of the patients had other diseases or syndromes known to affect tooth development.

Table 1. Key Characteristics of the Study Patients and Their Conditioning before SCT
DiagnosesNo. of patientsTBIHDCMean age at SCT (yrs) (range)
  • SCT: stem cell transplantation; TBI: total body irradiation; HDC: high dose chemotherapy; ARA-C: cytosine arabinoside at a dose of 3 g/m2 × 12; Cy: cyclophosphamide at a dose of 120–200 mg/kg; Mel: melphalan at a dose of 140–180 mg/m2; VMP: etoposide at a dose of 300 mg/m2, melphalan at a dose of 140 + 70 mg/m2, and cisplatin at dose of 90 mg/m2; ECT: etoposide at a dose of 250 mg/m2 × 3, carboplatin at a dose of 500 mg/m2 × 3, and triethylenethiophosphoramide at a dose of 300 mg/m2 × 3; NBL: neuroblastoma; ALL: acute lymphoblastic leukemia; AML: acute myeloid leukemia; CML: chronic myeloid leukemia; MDS: myelodysplastic syndrome; SAA: severe aplastic anemia RMS: rhabdomyosarcoma.

  • a

    One patient in the non-TBI group received total body irradiation excluding the head.

  • b

    The patient in the non-TBI group received total lymph node irradiation.

NBL18117a  41313.0 (1.0–7.9)
ALL12120723  4.9 (1.1–6.3)
AML880251  5.5 (2.2–9.0)
Wilms tumor505  5  5.4 (4.2–6.4)
CML22011   5.4 (5.2–5.5)
Lymphoma22011   6.1 (5.7–6.4)
MDS220 2   5.8 (2.2–9.4)
SAA211b 2   5.0 (3.9–6.1)
RMS101    14.7
Total5238141113131324.3 (1.0–9.4)

Informed consent was obtained from the patients and/or their guardians. The institutional review board of the Hospital for Children and Adolescents, University of Helsinki, approved the study protocol.

Treatment of the Patients before SCT

Multiagent chemotherapy of the patients with ALL, AML, and non-Hodgkin lymphoma proceeded according to Nordic protocols,48, 49 including prednisolone, vincristine, doxorubicin, methotrexate, L-asparaginase, cyclophosphamide, cytosine arabinoside, and 6-mercaptopurine. Conventional chemotherapy of patients with neuroblastoma (NBL) consisted of vincristine, cyclophosphamide, dacarbazine, cisplatin, and doxorubicin, added with ifosfamide and etoposide in individual cases.50 The five Wilms tumor patients and the RMS patient received vincristine, actinomycin D, cyclophosphamide, and doxorubicin. In addition, the patient with RMS received ifosfamide and etoposide. Hydroxyurea was the only chemotherapeutic agent given to the two patients with chronic myeloid leukemia, whereas patients with myelodysplastic syndrome and severe aplastic anemia did not receive any chemotherapy before SCT.

Three patients received fractionated cranial irradiation (CNS) in doses of 12 grays (Gy) to 2 patients with AML and 24 Gy to 1 patient with ALL. The amount of scattered irradiation in the CNS has been calculated to be 3–6% in different oral areas.14 Accordingly, 2 patients may have received scattered irradiation from approximately 0.4–0.7 Gy, and 1 patient from 0.7–1.4 Gy in the oral area. These 3 patients also received TBI (1 received 10 Gy and 2 received 12 Gy). Absorbed doses during TBI deviate from the reference dose in individual patients between −22.4% and +20.1% in various intraoral sites,51 which is much more than the amount of scattered irradiation in the CNS. Thus, the effect of minor scattered irradiation on developing teeth was considered to be insignificant. Local irradiation was given to 3 patients with NBL for skull metastases situated in the left frontal bone (20 Gy), right orbital area (20 Gy), and the left temporal bone (6 Gy). Due to shielding and good targeting of the radiotherapy, no scattered irradiation was delivered to the dental area. Two of these three patients also received TBI.

Conditioning for SCT

Chemotherapeutic agents with their dosages used for HDC before SCT are given in Table 1. In addition to HDC, the preparative regimen for SCT included TBI of 10–12 Gy, covering also the developing teeth, in 38 of 52 patients (TBI group). One patient with ALL received TBI of 10 Gy in a single fraction. All other patients received TBI in 2-Gy fractions. The remaining 14 patients belonged to the non-TBI group (Table 1).

Methods for Dental Examination

PRGs were taken at the Institute of Dentistry, University of Helsinki, except two that were obtained from a local health center. One author (P.H.) performed all radiographic assessments.

R/C ratios were calculated by measuring the root length (R) and the crown height (C) of all fully developed permanent teeth (microdontic teeth and third molars were excluded) from the PRGs and by dividing R by C. This method demonstrated good intraexaminer and interexaminer reproducibility in a previous study.47 For all assessed R/C ratios, individual SDSs, based on the tooth and gender-specific values in a healthy Finnish population,47 were calculated: SDS = (RCpat − RCmean)/SD, where RCpat means the R/C ratio of a patient's tooth, and RCmean and SD are the tooth and gender-specific mean R/C ratio and standard deviation of a corresponding tooth in a healthy population (see Fig. 1). The same two experienced radiographers had taken the PRGs of both the healthy individuals and the study patients. The SDS distribution of R/C ratios of fully developed permanent teeth (945 teeth of 52 patients) was examined. The mean number of measurable teeth per individual was 18.2, ranging from 1 to 28 (Table 2). For studying the possible effect of TBI and age at SCT on the R/C ratios, we divided patients and teeth into TBI and non-TBI groups and into three age groups. In the youngest group (Y), patients were age ≤ 3.0 years at SCT. In the middle group (M), patients were ages 3.1–5.0 years and in the oldest group (O), they were age ≥ 5.1 years at SCT. Subdivision of age groups to TBI and non-TBI groups was used in some analyses (Table 2).

Table 2. Number of Patients and the R/C Ratios Studied in Subgroups
GroupNo. of patientsNo. of teethMean no. (range) of R/C ratios measured per patient
  1. R/C: root/crown; TBI: total body irradiation; SCT: stem cell transplantation; Y: youngest group; M: middle group; O: oldest group.

All5294518.2 (1–28)
TBI3867417.7 (1–28)
Non-TBI1427119.4 (2–28)
Age at SCT   
Y) ≤ 3.0 yrs1422616.1 (1–25)
 TBI/non-TBI10/4154/7215.4/18.0 (1–23/2–25)
M) 3.1–5.0 yrs1426218.7 (10–28)
 TBI/non-TBI10/4185/7718.5/19.3 (10–28/10–28)
O) ≥ 5.1 yrs2445719.0 (4–28)
 TBI/non-TBI18/6355/12218.6/20.3 (4–28/4–28)

We also studied the percentage (number) of affected R/C ratios per individual. Variation in the number of fully developed permanent teeth per patient was high (range, 1–28 teeth). Only a few fully developed teeth in a patient do not necessarily give the right result as regards the ultimate injury in the dentition at the individual level. Thus, we formed a subgroup (n = 39) where we included only the patients who had > 50% of their theoretically measurable permanent teeth fully developed (SG39). The theoretic maximum per patient, when third molars and missing or microdontic teeth were excluded, varied from 16 to 28, and the number of the fully developed teeth from 11 to 28.

Statistical Analysis

The Statistical Package for the Social Sciences (SPSS for Windows), Version 10.0 (SPSS, Inc., Chicago, IL) was used in the statistical analyses. The statistical significance of continuous variables between the subgroups was tested with the Mann–Whitney U test. The Pearson chi-square test or Fisher exact test was used to test the statistical significance of category variables in the TBI and non-TBI groups and in the different age groups. P < 0.05 was considered to be significant.


  1. Top of page
  2. Abstract
  6. Acknowledgements

R/C Ratios of Permanent Teeth at the Level of Individual Patients

Altered root development was seen in all 52 pediatric recipients of SCT if R/C ratios outside ±2 SDSs were considered “abnormal.” Of the teeth measured per patient (range, 1–28 teeth; mean, 18.2 teeth), the number of affected teeth varied from 1 to 27 (mean, 13.9 teeth). In 90.4% (47 of 52) of the patients, R/C ratios deviating more than ±3 SDSs were measured. The number of teeth outside ±3 SDSs per patient ranged from 0–27 (mean, 10.2).

In SG39, the mean number (range) and percentage (range) of the fully developed permanent teeth studied per individual were 21.6 (11–28) and 87% (52–100%). In these teeth, the mean number (percentage) of 16.6 (77%) and 12.1 (56%) of the R/C ratios per patient were outside ±2 SDSs and ±3 SDSs, respectively (more details in Table 3). In 25 of 39 (64%) SCT recipients, > 50% of the teeth deviated > ±3 SDSs. Patients in the TBI group were more severely affected than the non-TBI patients: 24 of 29 TBI patients but only 1 of 10 non-TBI patients had > 50% of their teeth affected (P < 0.001). No such difference was present among the 3 age groups, although the mean number of teeth affected per patient and the number of patients having > 50% of their teeth injured tended to be higher in Group M (Table 3). In each age group, the patients who received TBI were more severely affected than the non-TBI patients. The differences were mostly significant (Table 3). The percentage of affected R/C ratios (outside ±2 SDS) per individual, in reference to TBI and age at SCT, is shown in Figure 2.

Table 3. Subgroup of 39 SCT Recipients with Advanced Tooth Developmenta
GroupNo. of patientsMean no. (range) of R/C ratios measured/patientMean no. of R/C ratios of ± 3 SDS (percentage of R/C ratios affected/patient, range)> 50% of R/C ratios of ± 3 SDS limits No. of patients (%)
  • Y: youngest group; M: middle group; O: oldest group.

  • a

    Data show number of R/C ratios studied, teeth with disturbed R/C ratios, and patients with > 50% of their teeth affected, in reference to the limits of ± 3 SD scores. Mann–Whitney U and chi-square (Fischer exact) test were used in the statistical analyses.

All3921.6 (11–28)12.1 (0–100)25 (64)
TBI2920.9 (11–28)14.1 (0–100)24 (83)
Non-TBI1023.7 (19–28)6.3 (0–85)1 (10)
Age at SCT    
Y) ≤ 3.0 yrs1119.6 (12–25)10.5 (0–100)8 (73)
M) 3.1–5.0 yrs1220.2 (11–28)15.2 (37–100)10 (83)
O) ≥ 5.1 yrs1624.1 (18–28)10.9 (0–100)7 (44)
Y) TBI/non-TBI8/318.1/23.3 (12–23/22–25)13.6/2.0 (50–100/0–22)8/0 (100/0)
M) TBI/non-TBI9/319.4/22.3 (11–28/19–28)16.1/12.3 (56–100/37–85)9/1 (100/33)
O) TBI/non-TBI12/423.8/25.0 (18–28/23–27)12.9/5.0 (0–100/0–38)7/0 (58/0)
TBI vs. non-TBI  P = 0.001P < 0.001
Y vs. M  P = 0.083P = 0.640
M vs. O  P = 0.099P = 0.054
Y vs. O  P = 0.902P = 0.239
Y) TBI vs. non-TBI  P = 0.013P = 0.006
M) TBI vs. non-TBI  P = 0.306P = 0.045
O) TBI vs. non-TBI  P = 0.045P = 0.088
thumbnail image

Figure 2. Percentage of teeth with root-crown (R/C) ratios deviating more than ±2 standard deviation scores in individual patients (n = 39) with advanced tooth development, in reference to total body irradiation (TBI) and age at stem cell transplantation (SCT).

Download figure to PowerPoint

R/C Ratios of Permanent Teeth at the Tooth Level

All 945 fully developed permanent teeth of 52 SCT patients were included in the R/C ratio analysis (microdontic teeth and third molars excluded) (Table 2). The mean SDS of all mature permanent teeth in SCT recipients was −3.4. In teeth exposed to a preparative regimen for SCT including TBI, the mean SDS was significantly smaller than in teeth exposed to HDC only, i.e., the mean SDS of R/C ratios were −4.0 and −1.8, respectively (P < 0.001) (Table 4).

Table 4. Mean SDS of R/C Ratios in 945 Fully Developed Permanent Teeth of 52 Patients Studied, Divided in Subgroups According to TBI and Age at SCTa
GroupMean SDS (SD) Significance    
  • SDS: standard deviation scores; SD: standard deviation; R/C: root crown; TBI: total body irradiation; SCT: stem cell transplantation; Y: youngest group; M: middle group; O: oldest group.

  • a

    The Mann–Whitney U test was used in the statistical analyses.

All teeth−3.4 (2.10)    
TBI−4.0 (1.87)    
Non-TBI−1.8 (1.83)    
 TBI vs. Non-TBIP < 0.001    
   Mean SDS (SD)  
Age at SCT  TBINon-TBITBI vs. Non-TBI
Y) ≤ 3.0 yrs−3.1 (2.69) −4.3 (1.81)−0.3 (2.14)P < 0.001
 Y vs. MP < 0.001Y vs. MP = 0.012P < 0.001 
M) 3.1–5.0 yrs−4.4 (1.98) −5.0 (1.94)−3.0 (1.27)P < 0.001
 M vs. OP < 0.001M vs. OP < 0.001P < 0.001 
O) ≥ 5.1 yrs−3.0 (1.60) −3.3 (1.58)−2.0 (1.22)P < 0.001
 Y vs. OP = 0.180Y vs. OP < 0.001P < 0.001 

Analyses of the 3 age groups (Y, M, and O) revealed that the mean SDSs at the time of the examination were −3.1, −4.4, and −3.0, respectively. The M group differed significantly from the others (P < 0.001) (Table 4). When patients in each age group were divided further in the TBI and non-TBI groups, the SDSs of R/C ratios deviated significantly more from the reference values in the TBI groups, indicating a poorer development of dental roots (Table 4). Teeth within the TBI and non-TBI groups and, separately, in the 3 age groups differed significantly from each other (Table 4), showing the most seriously affected mean SDSs of R/C ratios in the teeth of the patients who were 3.1–5.0 years old at SCT. The mildest changes were seen in the teeth of children in the O group if TBI was given, and in the Y group with HDC (Table 4). Median SDSs of R/C ratios with upper and lower quartiles and minimum and maximum values are shown in Figure 3.

thumbnail image

Figure 3. Median standard deviation score (SDS), upper and lower quartiles, and minimum-maximum values of root-crown (R/C) ratios of 945 permanent teeth in 52 stem cell transplant (SCT) recipients, in relation to total body irradiation (TBI) and age at SCT. The box represents the interquartile range containing 50% of the values. A line across the box indicates the median SDS of the R/C ratio. The whiskers show the highest and lowest values of SDS, excluding outliers. Outliers are marked with circles. For mean values of R/C ratios and significances between the groups, see Table 4.

Download figure to PowerPoint

In greater than three-fourths (77%) of the teeth measured, the R/C ratio was out of the normal area (i.e., outside ±2 SDSs) and in greater than one-half (56%) of the teeth, it deviated more than ±3 SDSs (Table 5). Significantly more R/C ratios of permanent teeth in the TBI group (85%) than in the non-TBI group (55%) were outside ±2 SDSs (P < 0.001). The percentages of R/C ratios deviating more than ±3 SDSs were 68% and 25%, respectively (P < 0.001) (Table 5).

Table 5. Number (Percentage) of R/C Ratios with SDSs Deviating more than ± 2 SDSs or ± 3 SDSs, with Minimum an Maximum SDS Values, in relation to TBI and Age at SCTa
 R/C outside ± 2 SDSs/teeth studied (%)R/C outside ± 3 SDSs/teeth studied (%)Minimum/maximum SDSs
  • R/C: root crown; SDS: standard deviation score; TBI: total body irradiation; SCT: stem cell transplantation; Y: younger group; M: middle group; O: oldest group.

  • a

    Chi-square test was used in the statistical analyses.

Patient group   
All725/945 (77)529/945 (56)−9.1/6.1
TBI575/674 (85)461/674 (68)−9.1/0.7
Non-TBI150/271 (55)681/271 (25)−6.2/6.1
Age group   
Y) ≤ 3.0 yrs160/226 (71)126/226 (56)−9.1/6.1
M) 3.1–5.0 yrs237/262 (91)191/262 (73)−9.1/−0.7
O) ≥ 5.1 yrs328/457 (72)212/457 (46)−8.8/2.1
TBI vs. non-TBIP < 0.001P < 0.001 
Y vs. MP < 0.001P < 0.001 
M vs. OP < 0.001P < 0.001 
Y vs. OP = 0.788P = 0.023 

The percentages of R/C ratios outside the normal ±2 SDS and ±3 SDS areas were highest in the M group. Teeth in the Y and O groups were affected equally, when divided according to ±2 SDS limits, and more affected in the Y group when ±3 SDS limits were used (P = 0.023). Differences among the 3 age groups are presented in Table 5. When age groups were subdivided according to TBI, the highest percentage of R/C ratios outside ±2 SDSs (i.e., 97%) was recorded in the teeth of patients who were ages 3.1–5.0 years at SCT and received TBI. In each of the 3 age groups (Y, M, and O), the number of the R/C ratios deviating more than ±2 SDSs and ±3 SDSs was significantly higher in the TBI groups than in the non-TBI groups (P < 0.001) (Figure 4).

thumbnail image

Figure 4. Percentage of the root-crown (R/C) ratios of 945 permanent teeth in 52 stem cell transplant (SCT) recipients outside normal ±2 SD and ± 3 SD scores (SDS) in 3 age groups (youngest [Y]: age ≤ 3.0 years at SCT; middle [M]: ages 3.1–5.0 years at SCT; oldest [O]: age ≥ 5.1 years at SCT), divided according to total body irradiation (TBI). In all comparisons, the differences between the TBI and non-TBI groups were significant (P < 0.001).

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  6. Acknowledgements

Our results indicate that disturbed root growth in permanent teeth regularly occurs after SCT, when performed in patients age < 10 years. HDC alone was surprisingly injurious but TBI further increased the dental damage. Although, after the treatment of childhood cancer, the worst dental consequences have been found in SCT patients, to our knowledge very little quantitative data concerning dental root development after SCT are available (i.e., reports on 16,19 18,25 and 7 patients23). In accordance with these earlier studies, the current study, including 945 teeth of 52 patients, revealed severe disturbances in the root growth of permanent teeth. All our patients were affected, and the R/C ratios were outside the normal ±2 SDS limits in 77% of the teeth studied, emphasizing the strong harmful effect of anticancer therapy on developing teeth. Root development, measured by the R/C ratio, was significantly worse in the teeth exposed to TBI compared with teeth not exposed to TBI. The most severely affected teeth were found in patients who were ages 3.1–5.0 years at the time of SCT.

Development of teeth and other epithelial appendages (e.g., hair and many exocrine glands) is a sequence of interactions between the surface epithelium and the underlying mesenchyme.52 Numerous signal molecules that are repeatedly used in the process carry on the tooth development from initiation through bud, cap, and bell stages to a fully developed tooth.53, 54 Any agent that is able to cause toxic death of odontogenic cells, inhibit their metabolic processes, or prevent or change signaling and cell communication is a potential cause of developmental aberrations. However, the molecular mechanisms mediating the toxicity of anticancer therapy on developing teeth have not been identified.

Nearly all SCT patients previously studied have received both HDC and TBI. Therefore, it has been impossible to distinguish between the separate roles of these treatments as regards dental defects.15, 17, 19 It has been reported, however, that TBI causes additive impairment when compared with HDC alone.15, 25, 55 Root surface area measurements have revealed the presence of small roots in mandibular teeth after pediatric anticancer therapy14, 19, 23 and especially after bone marrow transplantation.19, 23 In these earlier reports, all SCT recipients had received TBI, whereas we were able to study only 14 SCT patients without TBI. In accordance with earlier reports, TBI had a significant detrimental effect on root growth in the current study. A new finding was that HDC alone (non-TBI group) was very injurious and, unexpectedly, high percentages of the R/C ratios in the non-TBI teeth studied were outside ±2 SDSs (55%) and ±3 SDSs (25%). It is also interesting to note that all potential adverse effects of anticancer therapy on dental root development were not yet to be seen at the time of the study. Many teeth would require several years to complete their development to be assessable.

Age at diagnosis or at SCT in relation to dental adverse effects of pediatric anticancer therapy has been under study and young age at SCT has been detrimental to root development.15, 17, 19 Näsman et al.19 concluded that the younger the patient at SCT, the more detrimental the effects would be on root development. Our results are not fully in accord, as in our SCT group, the worst disturbances of R/C ratios were seen in patients ages 3.1–5.0 years at SCT, and patients age < 3 years were less affected. One difference between the two studies was that our study group also included non-TBI patients whereas all patients in the earlier report received TBI.19 This, however, did not explain the different results concerning the most vulnerable age of root development. According to our study, the R/C ratio was always most severely affected in the patients ages 3.1–5.0-years, regardless of whether TBI was included in the preparative regimen (Table 4). However, due to methodologic differences, direct comparison of these results is not possible. Näsman et al. used areas of mandibular teeth (third molars excluded) to calculate the crown-root ratio whereas we measured lengths of all fully developed permanent teeth for the assessment of the R/C ratio.

Only 3% of R/C ratios of mature permanent teeth in our patients who were ages 3.1–5.0 years at SCT and received TBI were within the normal limits of ±2 SDSs, whereas the percentage was 71% in the teeth of the youngest non-TBI patients (R/C ratios outside ±2 SDSs in Fig. 4). Thus, TBI was very detrimental in terms of root development, irrespective of age at SCT, although the most potent TBI effect on the R/C ratio occurred in the youngest age group (age ≤ 3.0 years at SCT) (Table 4; Figs. 3, 4).

The stage of tooth development, most closely connected to chronologic age, determines the fate of the tooth after an environmental insult. Our result indicating the poorest root growth after SCT at the ages of 3.1–5.0 years is logical. During that period, the crowns of the permanent incisors, canines, and first molars are completed and the root development begins.33 In these teeth, the proliferating cells, those most vulnerable to toxic insults, are located in the Hertwig epithelial root sheath and are unwanted targets of chemotherapy and/or irradiation. If the anticancer treatment is completed earlier, before the age of 3 years, the still undifferentiated cells in the region of the future dental root are more resistant to injury. Anticancer therapy delivered after 5 years of age is still able to disturb root growth, especially in late developing premolars and permanent second molars, but by that age, the roots in the early developing teeth have already reached a moderate length, improving the ultimate result.

The results of the current study prove that toxic or inhibitory effects of chemotherapeutic agents on odontogenesis, shown in histologic studies, may have clinical consequences in human teeth. Many questions still remain. It had been tempting to study which HDC protocols had the most severe effects on root growth, but the small number of patients and the wide variation in the SCT ages did not allow this analysis. Conventional chemotherapy, given in varying combinations to all but two patients before SCT, also may have had a role in disturbed root growth.

The major adverse effects of anticancer therapy to permanent dentition include agenesis, microdontia, and disturbed root development. Agenesis and microdontia after SCT have a considerable impact on dental development.55 The clinical significance of deficient root development is dependent on the number of teeth affected and the magnitude of root shortening. We believe the objective method used in the current study enabled us to document the late effects of anticancer therapy on the root growth of permanent teeth more accurately than was done in earlier studies. A high number of teeth per individual, especially if exposed to TBI, presented with deviating R/C ratios, as shown in our analysis of SG39 SCT recipients with advanced tooth development. When applying ±2 SDS limits, in 28 of 29 TBI patients, > 50% of their teeth had been affected and in 8 patients, damage had occurred in all permanent teeth. All 9 TBI patients ages 3.1–5.0 years at SCT had altered R/C ratios in 91–100% of their permanent teeth (Fig. 2). The teeth of the non-TBI patients were not protected, either. All patients were affected and an unexpectedly high number of patients (6 of 10 [60%]) showed the late effects of anticancer therapy in > 60% of their teeth (Fig. 2). Thus, the effect of HDC on dental root development was remarkable, and stronger than the effect of conventional chemotherapy, which affected 25–27% of the patients, as calculated from the earlier reports.17, 19

Minor root shortening most likely does not affect the life span of the tooth, if occlusion is acceptable and periodontal infections are avoided. Moderately shortened roots may be a risk factor for tooth loss. Short dental roots may also complicate orthodontic treatment or compromise the possibilities of prosthetic dentistry by reducing the ability of those teeth to give anchorage or carry masticatory forces. Loose teeth with extremely short roots may impair mastication ability and even change dietary habits. Early loss of such damaged teeth is also possible. The bony mass of alveolar ridges normally increases along with the dental root growth. Deficient root development results in diminished bone mass of alveolar ridges, impairing the possibilities for prosthetic rehabilitation. Accordingly, to minimize the clinical consequences of disturbed dental root development, intense and lifelong dental care is necessary for pediatric SCT survivors.


  1. Top of page
  2. Abstract
  6. Acknowledgements

The authors thank Jorma Torppa, M.Sc., for giving statistical advice, and Vesa Hölttä, M.Sc. (Tech.) for technical assistance with computers.


  1. Top of page
  2. Abstract
  6. Acknowledgements