PBIO 504 Literature Review Assignment Due Date: 04/11/2018
Identify an association of interest and select 3 – 5 most recent (published in the last 5-10 years) articles to review. Search PubMed (or any other peer-reviewed journal database) for these articles and download to read full text. Complete the literature review table below, list all references after the table, and write a one-page summary of the table.
Write a one-page report that summarizes the content of the table. This may include discussing about the use of different study designs, different study populations, different definitions for the outcome and for the main exposure, different statistical methods used for the analysis, and controlling for different sets of confounders.
Conclude with a short paragraph describing what you learned from this assignment.
In the “Outcome Measure” and “Main Exposure” (or main risk factor) column, include information on how the variable was classified (categorical vs. continuous), if it is self-reported, based on diagnosis, pathological results, etc., and cutoff points used (if applicable).
Association of interest: The potential of adverse effects due to the exposure to NSAID drugs (Nonasprin Nonsteroidal Anti-Inflammatory Drugs) in pregnant women that may or may not lead to spontaneous abortions.
Authors/Year Type of Study / Sample Size Population Characteristics Outcome Measure Main Exposure &
Other Variables Statistical Method Result
Daniel et.al. 2014 Historical cohort study.
This study was used to analyze the correlation between exposure to NSAIDS and spontaneous abortions. The study sample size was 65,457 pregnant women. The study included 65457 women ranging from ages of 15 and 45. The control group did not take NSAIDS, while the other group took NSAIDS. The year of pregnancy range from January 2003 to December 2009 The outcome measure was spontaneous abortions/miscarriages. The Databases of the Soroka Medical Center were used to compile information about the study’s population. The information complied included the pregnant women’s smoking status, medical status, demographics (race and age) and whether they took NSAIDS or not. There were 2 main exposure groups of NSAIDs: non-selective COX inhibitors (which included ibuprofen, naproxen and more) and COX-2 selective inhibitors, which included celecoxib, rofecoxib, and etoricoxib. For statistical analysis Fisher’s exact test was used for categorical variables and student t test was used for continuous variables. There wasn’t an association between NSAIDs and the risk of spontaneous abortions.
Nonselective Cox inhibitor data included an adjusted HR: 1.0 and 95% CI 0.99, 1.22
COX-2 selective inhibitor data included an adjusted HR: 1.43 and 95% CI 0.79, 2.59. However, there was an association between the use of indomethacin and spontaneous abortions during the first trimester. Indomethacin is used to delay premature deliveries and can cause a reverse causation bias.
Velez Edwards et.al 2012 A Non-clinical Cohort Study.
They used data from the RFTS study, a community-based pregnancy cohort, to examine NSAID use during pregnancy, specifically in the first trimester. There were 2,780 women who were enrolled in pregnancy cohort the study.
Participants of the study were pregnant women who were enrolled in the “Right from the Start” (RFTS) study within the years of 2004-2010. Participants of the study were 18 years old or older and had to have been pregnant or attempting to conceive for less than six months without reproductive technologies. The outcome measure was spontaneous abortions. Participants reported their outcomes, which was the verified through medical records. Gestational age was the variable use which falls under a continuous variable.
The participants reported their outcome and were interview after miscarrying, there is probable recall bias. The main exposure was NSAID use and no use. If NSAID were used, they were grouped by brand name, generic name and drug class. Researchers also took note of the participants age, height, weight, BMI, race, diabetes status, the number of time the participant has been pregnant, smoking status, induced abortion history and study site. To analyze this cohort study, researchers used Cox regression with STATA. To characterize the rate of spontaneous abortions to NSAID exposure and no NSAID exposure, gestational age was used. An alpha value of 0.05 significance level was used for all test. Of the 2,780 participants, there were 367 spontaneous abortions. The study concluded there is no association between NSAID and spontaneous abortions in the adjusted, HR: 1.00 CI 0.81, 1.23, and unadjusted models, HR: 1.01 CI 0.82, 1.24. Spontaneous abortions excluding salicylates, an ingredient in pain relieving medication, also concluded no association, unadjusted HR: .91 CI 0.72, 1.15 and adjusted HR: 0.91 CI 0.72, 1.14. Researchers were unable to analyze the use of prescribed NSAID on spontaneous abortions due to the small number of women who took prescribed NSAID.
Nakhai-Pour et.al. 2011 Nested Case-Control was the study of choice due to greater computational efficiency than cohort. The data of 4,705 women who had spontaneous abortions was collected from the Quebec Pregnancy Registry. 47050 controls were used. Participants were women 15-45 years old and insured by the RAMQ drug plan at least a year before their pregnancy and during their pregnancy. Exclusions included women who planned to have an abortion, had a spontaneous abortion after 20 weeks of pregnancy, or women used misoprostol, NSAID suppositories or known teratogens. during the first twenty weeks of pregnancy. The outcome measure was spontaneous abortions. The data of 4,705 women who had spontaneous abortions was collected from the Quebec Pregnancy Registry. The variables of this study were considered continuous variables. The main measure was the use of nonaspirin NSAIDs. Only nonaspirin NSAIDs that were reimbursed by the RAMQ drug plan were considered for this study.
To determine confounders variables such as sociodemographic, asthma, depression, use of other medications, and more were used. Logistic regression and odds ratios at a 95% CI were used to statistically analyze the data. The data was adjusted for confounders. Of the 4,705 women who had spontaneous abortions, 352 of them filled one or more nonaspirin NSAIDs while pregnant. The common NSAIDs used were naproxen, ibuprofen, rofecoxib, diclofenac, and celecoxib. The data found that there is an association between nonaspirin NSAID use and spontaneous abortions. The use of nonaspirin NSAIDs increase the risk of spontaneous abortions by 2.4-fold compared to no use of nonaspirin NSAIDS during pregnancy; OR:2.43 CI 2.12, 2.79.
List of References:
1. Daniel S, Koren G, Lunenfeld E, Bilenko N, Ratzon R, Levy A. Fetal exposure to nonsteroidal anti-inflammatory drugs and spontaneous abortions. CMAJ : Canadian Medical Association Journal. 2014;186(5):E177-E182. doi:10.1503/cmaj.130605.
2. Velez Edwards DR, Aldridge T, Baird DD, Funk MJ, Savitz DA, Hartmann KE. Periconceptional Over-the-Counter Nonsteroidal Anti-inflammatory Drug Exposure and Risk for Spontaneous Abortion. Obstetrics and gynecology. 2012;120(1):113-122. doi:10.1097/AOG.0b013e3182595671.
3. Nakhai-Pour HR, Broy P, Sheehy O, Bérard A. Use of nonaspirin nonsteroidal anti-inflammatory drugs during pregnancy and the risk of spontaneous abortion. CMAJ : Canadian Medical Association Journal. 2011;183(15):1713-1720. doi:10.1503/cmaj.110454.
Report:
The use of NSAIDs and spontaneous abortions is an essential study that should be heavily researched due to the common use of NSAIDs for pain, especially in pregnancies due to easy access to NSAIDs (over the counter) and the high pain experienced during pregnancies. One of the three studies, the study done by the Faculty of Pharmacy at University of Montreal3, concluded that NSAIDS in fact increased the chance of spontaneous abortions. However, the two other studies reported found there was no association between NSAIDs and spontaneous abortions. These results show that there needs to be more studies analyzing the correlation between NSAIDS and spontaneous abortions and further look into the safety of NSAIDS.
In the most recent study reported by the Canadian Medical Association, the criteria for the participants were said to be a good representation of women in the area, but this does not include women who do not have Kaiser Permanente or health insurance at all, which good be a substantial number of women. Indomethacin was mentioned in two studies but should have been mentioned in all studies since it is used to delay premature birth. In addition, a point of interest in the second reported study, they report mentioned recall bias as an important factor in all the studies that required interview post spontaneous abortions because the women may be depressed and feel guilt, which would cause them not to recall things that occurred in the pregnancy to the best of their ability2. This is extremely important in the analysis and credibility of the results of a study.
Overall due to the strong variation in a meta-analysis, it is evident more research should be done on the association of NSAIDS and spontaneous abortions. Although the majority of recent studies showed no, more research should be done to the commonality of over the counter NSAIDS and fragility of pregnancies. Women should have access to knowledge and proof of safe pain medicines that could be used during their pregnancies; very few studies create a concise fact on what treatments are better. While looking at the methods of the study I was glad I was able to navigate common statistical terms and also see the usage of STATA for statistical analysis in the studies. The usage of STATA and other tools for biostatics we learned could easily be applied in the future studies for NSAIDS and pregnancy safety.