This is the first study to explore pregnant women’s views on and adherence to LMWH therapy during pregnancy and the puerperium. We found women to adhere highly to the enoxaparin therapy during both the antenatal (97%) and postnatal (93%) periods, although in a third of women, adherence decreased by > 5% postnatally.
The high adherence rate observed in this study is perhaps not surprising. Colwell et al. , in the context of extended thromboprophylaxis with LMWH in orthopedic surgery, found that when properly instructed, patients can safely and efficiently self-administer the LMWH at home. This finding was also replicated by Watts et al. , who followed a small number of patients who had recently undergone low limb arthroplasty and prescribed LMWH for 6 weeks post-surgery. They found that of the patients who could potentially self-inject at home, 85% were compliant with their prophylactic therapy, and this has been further endorsed by the results of the recently published ETHOS registry . In a recent pilot study with the use of unfractionated heparin for second trimester placental insufficiency, the authors found no increase in maternal anxiety within a high-risk antenatal population attributable to regular self-administration of subcutaneous injections . McLintock et al.  are one of the few authors to explore the relationship between adherence to prescribed treatment and outcomes. In their study of high-risk pregnant women with mechanical prosthetic heart valves in-situ, of the women found to suffer from thrombotic complications whilst prescribed enoxaparin treatment, non-compliance and subtherapeutic anti-Xa levels were implicated in each case, with the authors concluding that good compliance with treatment was associated with a low risk of valve thrombosis and good fetal outcomes.
Others, however, have reported problems with self-administration of LMWH. Mengiardi et al.  followed 213 non-pregnant patients injecting LMWH in the community and found a self-reported non-compliance rate of 17.1%, with 38.9% of patients stating that self-administration of the injections required some effort. A similar finding was also reported by Spahn , who followed 207 patients following knee arthroscopy. Spahn reported problems with self-injection in 34.8% initially and 6.3% over the whole time.
In our study, although some women reported negative experiences when having to inject the enoxaparin, the majority appeared to cope with self-administration.
This study had higher adherence rates compared with the aforementioned studies, with apparently fewer problems. This may be in part due to the stronger beliefs about the necessity for treatment than concerns that the women had regarding the enoxaparin, as demonstrated by the necessity-concerns differential. Another key reason for the high adherence rate may be in part due to the regular monthly follow-up these women had in the hematology clinic. This follow-up provided an opportunity to discuss and check for any problems, and when necessary to reinforce the importance of the LMWH therapy during the antenatal period; this opportunity was lost to a certain degree during the postnatal period.
What is of importance to women?
One issue that came out from the questionnaire, which confirms our previous understanding, is that women like to know the precise teratogenic potential of a prescribed medicine. In this study, 96.8% of women wanted to know to what extent any medicine they took could affect their unborn baby as well as them, with many (78.5%) placing a higher priority on the impact the medicine has on the unborn baby, compared with any impact the medicine may have on them.
Of the women in the study, 91.4% stated that having to inject enoxaparin was not an issue for them, as long as their own and their unborn baby’s health was protected. When this question was repeated, just enquiring about their unborn baby’s health, 88.4% of women stated that having to inject enoxaparin was not an issue for them, as long as their unborn baby’s health was protected, illustrating women’s protective behavior regarding their unborn baby.
These findings suggest that although in the majority of cases, the women were prescribed enoxaparin for maintaining their own health, the fact that the women were pregnant, appeared to make them feel as if the enoxaparin was protecting their unborn baby’s health and maintaining their pregnancy. In some ways, such a belief is not completely wrong, because of the close physical relationship between a mother and child during the gravid period; however, if this belief continues during the postnatal period, then the women’s perceived necessity for enoxaparin may decline and explains in part the reason why postnatally, there was a group of women (suboptimal adherence group) in whom adherence to enoxaparin dropped by > 5% from their respective antenatal adherence rate. The necessity-concerns differential was significantly different between these two groups.
The drop in adherence in some women during the puerperium should perhaps not be surprising, as similar findings are reported in other clinical specialities. For example, in the field of HIV, as previously mentioned, adherence to HAART might increase antenatally, in an attempt by mothers to prevent vertical transmission, and then drop once the baby has been delivered [17–19].
When asked informally in clinic by the principal researcher on why they might have missed doses postnatally, two reasons were most commonly cited; firstly, the challenges of motherhood following birth meant that often they simply forgot because their regular routine had been disrupted (unintentional non-adherence). Others reported that following birth, their perceived necessity for the enoxaparin was less (intentional non-adherence); as their baby had been born successfully, they were less worried if they missed a dose. In their study of medication use in general amongst pregnant women, Sawicki et al.  reported the two most commonly cited reasons for women non-adhering to medication during pregnancy were forgetting to take medication and stopping medicine when they were feeling better. Though in Sawicki’s study, the reasons cited are by pregnant patients, clearly they resonate with our study population, as both unintentional and intentional non-adherence behaviors are exhibited by the women.
These finding do have important practice implications. In the current Royal College of Obstetric and Gynaecology guidelines for VTE prevention during pregnancy and the puerperium , the threshold for initiating prophylaxis is lower during the puerperium than antenatally. This means that women not eligible for enoxaparin therapy antenatally become eligible postnatally, for example women who have had an emergency Caesarean section (1 week prophylaxis), or women who are asymptomatic heterozygous carriers of Factor V Leiden, will be prescribed LMWH (6 weeks prophylaxis). Given our observations in this study and the fact that one of the reasons women appear to adhere to enoxaparin is that they feel it protects their unborn baby’s health, it might not be too inconceivable that women prescribed prophylaxis cold (i.e. just during the postnatal period) might not adhere to their treatment fully. Healthcare professionals, particularly those in contact with women during the early days following birth, requiring LMWH, should ensure patients have an opportunity to discuss prophylactic LMWH fully and are instructed on how to self-inject competently for those newly starting LMWH therapy and reinforce the necessity of adherence LMWH therapy for those who had already been injecting antenatally, so that women gain maximal benefit postnatally.
Additional questions asked in the questionnaire
The additional questions asked in the questionnaire were questions that the research team thought might be important factors for women injecting LMWH therapy during pregnancy and the puerperium. The key issues that came out from this section included the fact that one-third of women (33.4%) agreed with the statement which said that they sometimes worry that being on enoxaparin will limit the type of childbirth delivery they could have, with a similar number (31.8%) being worried that being on enoxaparin will limit the type of pain relief they could have at the time of childbirth.
Most of the women recruited into this study were asked to complete the questionnaire within a month or two of starting LMWH therapy. In our clinic, discussions about childbirth and options for pain relief are bought to the forefront during the third trimester. Our findings suggest that briefly discussing childbirth and pain relief issues with women on antenatal LMWH early on in their pregnancy might overcome any anxieties that some women may have, particularly as they can be managed.
This study is limited by the fact that the questionnaire was only administered once during the women’s pregnancy. Following the result that postnatal adherence may fall in some women, it would have been interesting to assess how women’s views may have changed as pregnancy progresses and following delivery; this might form the basis of future studies. It must be borne in mind that this study was part of a wider pharmacokinetic modelling study of enoxaparin during the antenatal period. Therefore, the antenatal adherence rates we report might represent an overestimate of the true adherence rate in the real world, due to the fact that women knew enoxaparin (anti-Xa activity) was being monitored. Our clinical experience, however, suggests that women requiring LMWH in this setting are often highly motivated, as they wish for a successful obstetric outcome. Due to the pharmacokinetic nature of the wider study, we had to exclude women from the study who were non-adherent to enoxaparin. During the course of this study, we found this to be the case in three patients and so the questionnaire was not administered to them. It would have been interesting to see what their responses would have been to the questionnaire and assess how they differed from those of the cohort of women who adhered to treatment. Finally, though it was emphasized to the women participating that their responses would remain confidential and not shared with the clinical team, it’s impossible to exclude the fact that some women may have responded with socially acceptable responses.