Association between maternal age at conception and risk of idiopathic clubfoot: A meta-analysis

MethodsWe searched PubMed-MEDLINE, EMBASE, and the Cochrane Library up to June 2015 and supplemented the search with manual searches of the reference lists of the articles identified. 11 studies published between 1990 and 2015 were pooled. We investigated heterogeneity in maternal age and whether publication bias might have affected the results.ResultsCompared to a control group, maternal age at conception of between 20 and 24 years old was associated with an increased risk of occurrence of clubfoot (OR = 1.2, 95% CI: 1.1–1.4). No such association was found for the age groups of ≥ 35, 30–34, 25–29, and < 20 years. There was no heterogeneity in the age groups of ≥ 35, 30–34, and 20–24 years, moderate heterogeneity in the 25- to 29-year age group, and a large degree of heterogeneity in the group that was < 20 years of age. The prediction intervals for the age groups of 25–29 and < 20 years were 0.56 to 1.3 and −0.39 to 2.4, respectively. We found no evidence of significant publication bias.InterpretationFrom the results of this meta-analysis of 11 studies, maternal age at conception between 20 to 24 years of age appears to be associated with an increased risk of occurrence of clubfoot.StatisticsFor the studies included, we determined the pooled OR with 95% confidence interval (CI) for the maternal age groups versus control groups. Statistical heterogeneity was evaluated with the I2 statistic (Higgins and Thompson 2002): I² > 50% was considered statistically significant (Hedges and Pigott 2001). Here, we chose a random effects model to calculate the pooled OR and CI for more conservative results. If there was a moderate or large degree of heterogeneity between the studies included (with I² > 25% as a guide (Higgins et al. 2003)), we calculated a prediction interval considering the potential effect within an individual study setting (Riley et al. 2011). We used visual inspection of funnel plots, Egger’s test, and Begg’s test to determine whether publication bias might have affected the statistical results (Begg et al. 1994, Egger and Smith 1998). All statistical analyses were conducted using Revman 5.3.5 (The Nordic Cochrane Centre, Copenhagen, Denmark) and Stata 12.0. Any p-value of < 0.05 was considered statistically significant.DiscussionWhether or not maternal age is associated with the risk of clubfoot has been unlear. A recent population-based case-matched control study in Hungary from Csermely et al. (2015) showed a higher proportion of cases of clubfoot in offspring of mothers in the youngest age group (≤ 19 years), and a borderline excess of cases in offspring of mothers of older age (i.e. ≥ 35 years). Palma et al. (2013) reported an association between maternal age (of < 23 years) at conception and increased risk of clubfoot. Parker et al. (2009) also found that maternal age (of < 23 years) was associated with clubfoot. Similarly, a study by Nguyen et al. (2012) showed that young maternal age (of < 25 years old) was associated with increased risk of clubfoot. Dickinson et al. (2008) found a trend of increasing risk of clubfoot with decreasing maternal age, with women in the youngest age group (< 20 years) having the highest risk (OR = 1.6, 95% CI: 1.1–2.2) relative to women aged 30 or more. Multivariate analysis by Mahan et al. (2014) revealed that the strongest predictor in prenatal detection was a maternal age of ≥ 35 years (OR = 3.5). 1 study showed a negative correlation between occurrence of clubfoot and maternal age of > 35 years (Kancherla et al. 2010). Hollier et al. (2000) reported a higher risk of clubfoot, diaphragmatic hernia, and cardiac defects in mothers of older age. However, the opposite results—with no association between maternal age at conception and risk of clubfoot—have also been reported (Alderman et al. 1991, Cardy et al. 2007, Cardy et al. 2011, Honein et al. 2000, Moorthi et al. 2005, Sahin et al. 2013).The findings from this meta-analysis of 11 case-control studies were that a maternal age of 20–24 years old at conception was statistically significantly associated with increased incidence of clubfoot. There was a 20% higher risk of clubfoot than in the control group. We found no evidence that other maternal age groups were associated with an increased risk of clubfoot. We found no heterogeneity in maternal age groups of 35 years or more, 30–34 years, or 20–24 years, moderate heterogeneity in the 25- to 29-year age group, and a large degree of heterogeneity in the age group of less than 20 years. The prediction intervals for the age groups of 25–29 years and less than 20 years both overlapped the OR value of 1. After sensitivity analysis of these groups, the same results were observed, indicating that our meta-analysis was relatively stable. The source of heterogeneity in these 2 maternal age groups might be traced to different survey regions, racial variation, and confounding factors.A strength of our meta-analysis was the large number of cases included (n = 15,242 in the clubfoot group and 97,041 in the control group). One limitation might be the possible effects of having only a small number of studies, as only 11 studies were included. We did not find any publication bias because of such effects, but Egger’s test is known to have low power when less than 20 studies are included in a meta-analysis (Sterne et al. 2000). By not adequately controlling for confounders, our findings may have been biased in either direction (i.e. an exaggeration or underestimation of the risk estimate).A large degree of heterogeneity was found for the maternal age group of less than 20 years (I² = 74%), but we did not conduct a subgroup analysis using the limited information from crude data provided by the studies included.The risk of clubfoot may be attributed to sociodemographic factors, socioeconomic status, education status, social culture, and other factors. A possible explanation of why a maternal age of between 20 and 24 years of age would be associated with an increased incidence of clubfoot might be the first-born baby boom period in this age group. Epidemiological studies have consistently found a higher prevalence of idiopathic clubfoot in primiparous mothers (Honein et al. 2000, Skelly et al. 2002, Carey et al. 2005). Our meta-analysis suggests that there may be a negative correlation between the risk of clubfoot and higher maternal age. The reason may be attributed to the birth of another baby in a family, which leads to a decreased clubfoot risk at 25 or older years of maternal age. It has been suggested that the higher the parity of a pregnant woman, the lower the risk of clubfoot in the baby (Carey et al. 2005, Dickinson et al. 2008, Kancherla et al. 2010). The low risk of congenital clubfoot in mothers less than 20 years old may be explained by the higher incidence of abortion and miscarriage in this age group.

Supplementary data

Tables 1 and 2 and Figure 2 are available on Acta’s website (www.actaorthop.org), identification number 9564.YBL and JD contributed substantially to the acquisition of data and drafted the article. JZ and ZKW analyzed and interpreted the extracted data. LZ contributed to the conception and design of the article. HL and XY conducted the literature review and extracted the data from the studies that were included. CLX was responsible for statistical testing.No competing interests declared.     

Outcome of total hip arthroplasty,but not of total knee arthroplasty,is related to the preoperative radiographic severity of osteoarthritis: A prospective cohort study of 573 patients

ResultsAdjusted for sex, age, preoperative scores, BMI, and Charnley score, radiographic severity of OA in THA was associated with improvement in HOOS “Activities of daily living”, “Pain”, and “Symptoms”, and SF36 physical component summary (“PCS”) scale. In TKA, we found no such associations.InterpretationThe decrease in pain and improvement in function in THA patients, but not in TKA patients, was positively associated with the preoperative radiographic severity of OA.DiscussionThis prospective study in patients undergoing THA and TKA showed that changes in scores over time were greater in patients with more severe radiographic OA. The difference was statistically significant for a number of clinical outcomes in THA patients, but not in TKA patients.Overall, our results are in line with the literature, with the majority of studies concluding that more severe radiographic OA preoperatively is associated with better outcomes in THA or TKA (Dowsey et al. 2012, Valdes et al. 2012, Keurentjes et al. 2013). Concerning THA specifically, similar to the present study, Valdes et al. (2012) reported greater improvements in pain 3 years after surgery in patients with severe radiographic OA preoperatively. Greater improvements in the SF subscale and summary scale scores were seen in patients with higher KL scores in a study by Keurentjes et al. (2013), but the differences were not statisticaly significant.Regarding TKA, our study did not show any statistically significant differences between the outcomes in patients with different grades of radiographic severity, although—as in the study by Cushnaghan et al. (2009)—greater improvements were generally seen in patients with higher KL grades. In contrast, Valdes et al. (2012) and Keurentjes et al. (2013) found statistically significantly better outcomes in TKA patients with severe radiographic OA, and similar results were seen in some of the analyses in the study by Dowsey et al. (2012). Comparisons with the literature are, however, hampered by the large diversity in study designs and analyses.It is difficult to draw conclusions about the clinical relevance of the results of our study and of previous ones. Firstly, there are several factors associated with worse outcomes after THA/TKA, such as older age, female sex, obesity, worse general health, involvement of other joints, and a lower level of education (Dieppe et al. 2009, Gossec et al. 2011). Only from large, prospective studies using a standardized set of preoperative characteristics and outcome assessments done at fixed time points can true prediction models including all potentially relevant determinants be derived, which afterwards need to be validated in multiple settings and countries. However, we can interpret the absolute change scores as observed in the different groups according to radiographic severity. A recent systematic review by Keurentjes et al. (2012) found that overall minimally clinically important differences (MICDs) in HRQoL in THA/TKA have limited precision and are not validated using external criteria. The study which is most comparable to our study is that from Clement et al. (2014). In that study, the MCID in OKS for the difference between preoperatively and 1 year postoperatively was 15.5 (95% CI: 14.7–16.4). In our study, generally patients in both the mild and severe OA groups achieved this improvement, indicating that the clinical relevance of a statistically significant difference may be limited.A main strength of our study was the inclusion of a wide range of validated PROMs, covering all items of disease-specific outcome measures in functioning, pain, and health-related quality of life. Using all these outcome measures, both measures of pain and daily activities, we observed differences between groups according to radiographic severity. Another strength was that all radiographs were read by a single observer with extensive experience, who was blinded regarding patient data. In addition, this was a prospective study with a relatively large cohort with only 20% loss to follow-up in the THA group and only 23% loss to follow-up in the TKA group.Our study also had a number of limitations. It only included KL grading applied to the anteroposterior and posteranterior radiographs from the preoperative hip and knee.In the study by Dowsey et al. (2012), not only KL grading but also the severity of joint space narrowing (JSN; 0–3) and osteophyte formation (0–3) using the Osteoarthritis Research Society International (OARSI) atlas, and the degree of bone attrition, were taken into account. In that study, radiographs showing advanced OA (KL 3–4) were further subdivided by including data from the individual score of JSN and bone attrition.In addition, the patients included in the present study were a selection of all patients who underwent THA or TKA and it was carried out in 1 center in 1 country. However, the preoperative characteristics of the patients and their change scores over time are well in line with those observed in other large cohorts (Nilsdotter et al. 2003, Dieppe et al. 2009, Beswick et al. 2012).In conclusion, this study shows that in patients who underwent THA, but not TKA, more severe radiographic OA preoperatively was associated with a better outcome regarding pain and function.

Supplementary data

Tables 1 and 2 are available on the Acta Orthopaedica website, www.actaorthopaedica.org, identification number 8277.CT, MF, and TPMVV: conception and design, analysis and interpretation of the data, drafting of the article, provision of study materials or patients, statistical expertise, and collection and assembly of data. MJH, RLT, HMK, CSL, and SHM: provision of study materials or patients, administrative, technical, or logistic support, and collection and assembly of data. HMK and RGHHN: critical revision of the article, statistical expertise.This study was supported by the Dutch Arthritis Association (grant number LLP13).No competing interests declared.     

Missed low-grade infection in suspected aseptic loosening has no consequences for the survival of total hip arthroplasty: 173 patients followed for 6 to 9 years

Background and purpose — Aseptic loosening and infection are 2 of the most common causes of revision of hip implants. Antibiotic prophylaxis reduces not only the rate of revision due to infection but also the rate of revision due to aseptic loosening. This suggests under-diagnosis of infections in patients with presumed aseptic loosening and indicates that current diagnostic tools are suboptimal. In a previous multicenter study on 176 patients undergoing revision of a total hip arthroplasty due to presumed aseptic loosening, optimized diagnostics revealed that 4–13% of the patients had a low-grade infection. These infections were not treated as such, and in the current follow-up study the effect on mid- to long-term implant survival was investigated.Patients and methods — Patients were sent a 2-part questionnaire. Part A requested information about possible re-revisions of their total hip arthroplasty. Part B consisted of 3 patient-related outcome measure questionnaires (EQ5D, Oxford hip score, and visual analog scale for pain). Additional information was retrieved from the medical records. The group of patients found to have a low-grade infection was compared to those with aseptic loosening.Results — 173 of 176 patients from the original study were included. In the follow-up time between the revision surgery and the current study (mean 7.5 years), 31 patients had died. No statistically significant difference in the number of re-revisions was found between the infection group (2 out of 21) and the aseptic loosening group (13 out of 152); nor was there any significant difference in the time to re-revision. Quality of life, function, and pain were similar between the groups, but only 99 (57%) of the patients returned part B.Interpretation — Under-diagnosis of low-grade infection in conjunction with presumed aseptic revision of total hip arthroplasty may not affect implant survival.Aseptic loosening and infection are 2 common causes of revision in total hip arthroplasty (THA) (Sadoghi et al. 2013). Data from the Norwegian Arthroplasty Register and the Swedish Hip Arthroplasty Register have shown that antibiotic prophylaxis, administered either systemically, locally, or combined, prevents infection and thus reduces revision rates due to infection. Interestingly, it has also been shown that the rates of revision due to aseptic loosening decrease with the use of antibiotic prophylaxis (Malchau et al. 2002, Engesaeter et al. 2003). As theoretically the use of antibiotics should not have any influence on aseptic loosening, this suggests that the diagnosis of infection was inadequate. Under-diagnosis of infection in THA could possibly reduce the survival of revision hip implants.One of the major challenges when diagnosing low-grade infection is the accurate identification of microorganisms. These can be difficult to detect with routine diagnostics because of previous antimicrobial exposure or the requirement of certain microorganisms for specific nutrients, and also possibly due to reduced growth rates of biofilm-residing organisms (Fux et al. 2003, Schafer et al. 2008). To overcome this obstacle, new techniques for detection and identification of bacteria have been introduced. For example, polymerase chain reaction (PCR) on bacterial 16S ribosomal RNA (rRNA) can theoretically detect as little as 1 bacterium in a sample (Trampuz et al. 2003, Clarke et al. 2004, Fenollar et al. 2006, Kobayashi et al. 2009, Bergin et al. 2010). The apparent disadvantage of this feature is that PCR detection is susceptible to bacterial contamination (Panousis et al. 2005).Previously, our group developed and validated a combined 16S rRNA PCR and reverse line blot hybridization (RLBH) technique, which could identify many bacteria at the species level (Moojen et al. 2007). This combined technique was then used in a clinical study to test the hypothesis that there is under-diagnosis of infection in patients undergoing a THA revision due to aseptic loosening. In 7 Dutch hospitals, 176 patients undergoing revision of their THA following a preoperative diagnosis of aseptic loosening were included. During surgery, tissue biopsies were obtained for microbiological examination, pathological analysis, and broad-range 16S rRNA PCR with RLBH. We showed that 7 (4%) of these patients had a bacterial infection and an additional 15 (9%) were suspected of having an infection. Of these 22 patients, 2 were given a prolonged period of treatment with antibiotics. After a 1-year follow-up of 170 of the 176 patients, none of the 22 patients with an infection or suspected infection had undergone additional surgery.In the current study, we performed a mid- to long-term follow-up on this cohort to investigate the effects of missed low-grade infection on implant survival and clinical outcome. The hypothesis was that patients with an undiagnosed low-grade infection would show a higher rate of implant failure and poorer clinical outcome.     

Antibiotic cement was associated with half the risk of re-revision in 1,154 aseptic revision total knee arthroplasties

ResultsThere were 114 re-revisions (10%) with a median time to reoperation of 3.6 years (interquartile range (IQR): 2.6–5.2). The infection rate was 2.9% (34/1,154) and accounted for 30% of re-revisions (34 of 114). In adjusted models, use of antibiotic-loaded cement was associated with a 50% lower risk of all-cause re-revision surgery (hazard ratio (HR) = 0.5, 95% CI: 0.3–0.9), age with a 20% lower risk for every 10-year increase (HR = 0.8, CI: 0.7–1.0), body mass index (BMI) with a 20% lower risk for every 5-unit increase (HR = 0.8, CI: 0.7–1.0), and a surgeon’s greater cumulative experience (≥ 20 cases vs. < 20 cases) with a 3 times higher risk of re-revision (HR = 2.8, CI: 1.5–5).InterpretationRevised TKAs were at high risk of subsequent failure. The use of antibiotic-loaded cement, higher age, and higher BMI were associated with lower risk of further revision whereas a higher degree of surgeon experience was associated with higher risk.ResultsThe study cohort consisted of 1,154 aseptic revision TKAs. The mean age of the cohort was 65 (SD 10) years old, 61% were female, 32% were diabetic, 64% were white, 28% had a BMI greater than 35, and 52% had an ASA score of < 3. (2006a, Mortazavi et al. 2011, Bae et al. 2013, Luque et al. 2014). Sierra et al. (2004) reported a 40% cumulative revision risk at 20 years in 1,814 cases operated over a 30-year period. The Finnish Arthroplasty Register reported 79% survivorship of revision TKA at 10 years in 2,637 revision TKAs (Sheng et al. 2006b). In a smaller and more recent study by Luque et al. (2014), 125 aseptic revisions were reported with a minimum follow-up of 7 years and an 8-year survival of 88%. The causes of revision in our cohort parallel those presented in other studies where infection, aseptic loosening, and pain due to instability or stiffness consistently remain the leading causes of revision (Sheng et al. 2006a, Mortazavi et al. 2011, Bae et al. 2013, Luque et al. 2014).We found that the use of antibiotic cement at the time of the index revision was associated with half the risk of future all-cause revision. In a recent randomized controlled trial, the effect of vancomycin-loaded cement use in the context of 183 low-risk, aseptic revision TKAs was evaluated and a statistically significant reduction in postoperative deep infections at a minimum follow-up of 36 months was reported (none in the intervention group became infected, as compared to 7% in the control group) (Chiu and Lin 2009). However, several studies that have evaluated the association between antibiotic-loaded cement and infection after primary TKA surgery have arrived at inconsistent results. A review did not find antibiotic-loaded cement to be consistently associated with a lower risk of infection in modern, primary TKA (Jiranek et al. 2006). Also, a study by Namba et al. (2013), using the same data source as in our study, found that—paradoxically—antibiotic-loaded cement was associated with a slightly higher risk of surgical site infection after TKA. The higher risk of infection in revision TKA than in primary TKA procedures is probably the reason why we identified such a substantially lower risk of re-revision surgery in cases where antibiotic-loaded cement was used. Furthermore, the use of antibiotic bone cement in cases of subclinical or undiagnosed infections might favorably affect the results of the procedure.A second factor, the surgeon’s cumulative experience at the time of the index revision, was associated with a higher risk of re-revision surgery. As the most complex and high-risk cases are referred to more experienced surgeons, we believe that this finding is probably a proxy for case complexity, which is something we could not adjust for in our analysis. To our knowledge, the finding that higher BMI was associated with a small but statistically significantly lower risk of revision has not been reported elsewhere with respect to outcomes of revision TKA surgery, while the decrease in risk with older age has (Sheng et al. 2006b). We can only infer that activity levels may be lower in older patients or in those with a higher BMI, and that a combination of higher morbidity, higher perceived risk, and lower demand may lead to a lower revision risk associated with increasing age (Sheng et al. 2006b).After adjusting for all other risk factors, we did not find sex, race, ASA score, diabetic status, surgeon volume, hospital volume, surgeon’s TJA fellowship training, or use of hinged prosthesis at index revision to be associated with the risk of re-revision surgery.Our study had several limitations and strengths. Among the limitations, some of the data sourced for this study required voluntary surgeon participation (currently at 95%) with non-differential rates of participation across sites. There were missing data, but they were handled in the statistical analysis using multiple imputations. We do not feel that either of these limitations would affect outcomes. In addition, due to our sample size, which limited by the number of factors that could be evaluated at this time, in our analysis we were not able to evaluate the influence of surgical factors such as fixation method (i.e. cemented, uncemented, or hybrid), the extent of the index revision (i.e. 1, 2, or more components revised) and structural issues such as bone quality. Doing this might identify other risk factors for early revision. Furthermore, our decision to limit the cohort to those patients for whom the primary procedure had been captured in the registry limited us to a short follow-up period. Longer follow-up might have shown a higher percentage of patients revised for component wear or loosening. It is also likely that, as with any study of revision TKA, some patients in the cohort may have had an undiagnosed low-grade infection and that this might have skewed the overall risk of infection. Regarding surgeon experience, we note that the results can only reflect the period of data collection for the study and not lifetime experience.Among the strengths of the present study, we can include the large number of cases treated across multiple medical centers in a community-based setting, which should have provided data comparable to the experience of the majority of community surgeons. Furthermore, there was only a small possibility of data-handling bias due to the use of our integrated electronic medical record. Additionally, all of the outcomes evaluated in this study were manually adjudicated by a trained research assistant to guarantee the accuracy and integrity of the information reported, thus ensuring the high internal validity of the information reported.In summary, the most striking finding from our study of 1,154 aseptic TKA revisions is that the use of antibiotic-loaded cement was associated with half the risk of subsequent revision surgery. Infection, instability, pain, and aseptic loosening remain ongoing challenges associated with a 20% cumulative probability of failure at 5 years. Surgeons and patients alike must be cognizant of the potential for poor long-term outcomes following revision TKA.All the authors contributed to the study design and contributed substantially to collecting the data, interpreting the results, drafting the article, and to revision. PHC and MCSI conducted the statistical analysis.No competing interests declared.     

Maximum temperatures of 89°C recorded during the mechanical preparation of 35 femoral heads for resurfacing

Background and purpose

We noticed that our instruments were often too hot to touch after preparing the femoral head for resurfacing, and questioned whether the heat generated could exceed temperatures known to cause osteonecrosis.

Patients and methods

Using an infra-red thermal imaging camera, we measured real-time femoral head temperatures during femoral head reaming in 35 patients undergoing resurfacing hip arthroplasty. 7 patients received an ASR, 8 received a Cormet, and 20 received a Birmingham resurfacing arthroplasty.

Results

The maximum temperature recorded was 89°C. The temperature exceeded 47°C in 28 patients and 70°C in 11. The mean duration of most stages of head preparation was less than 1 min. The mean time exceeded 1 min only on peripheral head reaming of the ASR system. At temperatures lower than 47°C, only 2 femoral heads were exposed long enough to cause osteonecrosis. The highest mean maximum temperatures recorded were 54°C when the proximal femoral head was resected with an oscillating saw and 47°C during peripheral reaming with the crown drill. The modified new Birmingham resurfacing proximal femoral head reamer substantially reduced the maximum temperatures generated. Lavage reduced temperatures to a mean of 18°C.

Interpretation

11 patients were subjected to temperatures sufficient to cause osteonecrosis secondary to thermal insult, regardless of the duration of reaming. In 2 cases only, the length of reaming was long enough to induce damage at lower temperatures. Lavage and sharp instruments should reduce the risk of thermal insult during hip resurfacing.Hip resurfacing can fail due to osteonecrosis (Amstutz et al. 2004, Daniel et al. 2004). Osteonecrosis has been explored by surrogate means. The femoral head is devascularized by the posterior approach (Steffen et al. 2005, Beaule et al. 2006, Khan et al. 2007) and its blood flow is reduced by 50% if the neck is notched (Beaule et al. 2006). Temperatures during femoral head preparation are unknown and could be a cause of osteonecrosis. Temperatures may reach 68°C when cement is polymerizing during resurfacing (Gill et al. 2007).The effect of heat generated in bone at the cellular level is difficult to quantify. The important factors are the peak temperature and the duration of the thermal insult. With higher temperatures, a shorter exposure is needed to cause injury (Lundskog 1972, Berman et al. 1984). Thermal insult of 47°C for 60 s is the threshold for bone injury (Ericksson and Albrektsson 1983). Exposure to 50°C for 30 s causes widespread injury to bone 1 mm from the point of exposure (Lundskog 1972) and 55°C for 1 min causes marrow necrosis (Berman et al. 1984). Bone alkaline phosphatase is denatured at 56°C (Posen et al. 1965). When bone reaches to a temperature of 70°C or more, macroscopic bone necrosis can be seen intraoperatively. Cell necrosis occurs at temperatures of 70°C within 1 s (Moritz and Henriques 1947). There is histological evidence of bone necrosis after exposure to 70°C for 1 min (Berman et al. 1984) and 80°C for 5 s (Lundskog 1972).We noticed that our instruments were often too hot to touch after preparing the femoral head for resurfacing and wondered whether the heat generated during femoral head preparation might exceed the temperatures known to cause osteonecrosis.