New PCOS Ovulation Patterns Identified

Women with Polycystic Ovarian Syndrome (PCOS) almost always have irregular menstrual cycles. This can make predicting ovulation and achieving pregnancy challenging. New home technology such as OvuSense is making monitoring your cycle, sharing it with your doctor, and making decisions based on real data and informed interpretation a reality. Research using this fresh technology is offering promising results and, in many ways, proving that there is no “normal cycle.” In fact, new research indicates that there are at least 3 PCOS ovulation patterns that may have real implications on fertility treatment.

What makes this PCOS cycle identification possible?

Smartphone technology is at the forefront of a diagnostics revolution. With devices collecting, processing and sharing large volumes of data in real time. Previously, this data collection would have been the labour-intensive result of a long hospital stay. Nowadays, we are seeing another dimension of smartphone capability with smart analysis of data at the point of collection in a user’s home. This is likely to transform technology when allied with emerging data processing technologies such as Artificial Intelligence (AI).

This transformation is well applied by home ovulation monitoring, for which there are now a number of options using smartphones. These monitoring systems are typically purchased by women looking to select the most viable date and time for conception. Several, including OvuSense, are achieving a high degree of accuracy for those with PCOS.

Putting technology to work in research

A 2019 study was presented at the American Society of Reproductive Medicine (ASRM) Congress. In this study, the researchers set out to determine if averaged nocturnal vaginal Core Body Temperature (CBT) measurements recorded during non-menstruation could describe atypical patterns potentially associated with reduced fertility¹.

The researchers sampled CBT readings from 10,463 ovulatory cycles provided by 6,647 users. The female age ranged from 20 to 52 years ,with cycle length 11 to 190 days (90% 22 to 47 days). For this purpose, the OvuSense^TM system was utilised^2,3.

Easy to use technology make detailed observation possible

OvuSense^TM is a fully approved Class II medical device for measuring CBT, allowing patients to monitor their ovulation patterns. It utilises a temperature sensor placed in the vagina overnight, to measure CBT with a resolution of 0.003 degrees Celsius. The data from the sensor is downloaded each morning to the dedicated OvuSense^TM app.

The system makes use of a proprietary moving-average calculation to produce a smooth CBT analysis curve from CBT readings taken every 5 minutes. In particular, the researchers studied the proportion of normal and atypical CBT patterns, the frequency of their occurrence, and associations between different patterns.

What information was gleaned from the data collected?

3 Atypical CBT patterns observed

The researchers found that, of the 10,463 ovulatory cycles observed, 3,721 cycles exhibited one or more novel atypical core body temperature patterns. They suggest that it is likely that continuous vaginal temp patterns closely reflect luteal progesterone changes, hence describing subtle progesterone secretion or metabolism anomalies, which had not previously been recognized.

Atypical CBT patterns tended to fall into three categories:

1. Crash to baseline

A fall of more than 0.2 degrees Celsius to the lowest averaged CBT point in the cycle (baseline). This was observed in 1,481 cycles (14.2%) in 1,352 users (20.3%).

Here is an example (page: 37. Length of cycle: 26 days. Ovulation on Day 22)

Blue line: shows the best representative raw CBT value produced by the OvuSenseTM algorithm for each set of overnight measurements (taken every 5 minutes).

Blue shading: OvuSense^TM– detected day of ovulation.

Green line: this is the smooth weighted average CBT curve as used by the OvuSense^TM algorithm to predict ovulation up to 24 hours in advance using this current cycle’s data and then confirm ovulation.

Green shading: the ovulation window

Grey line:  the typical pattern, which might have been expected for this cycle, taking into account an expected textbook middle of the cycle ovulation.

Possible explanation:

Researchers postulate that this anomaly could possibly be due to high progesterone levels early in the cycle. One of the conditions that could cause this might be polycystic ovary syndrome (PCOS) – a leading cause of infertility.

2) False start

A false start pattern was seen when a rise of more than 0.1 0 Celsius did not result in ovulation, but instead initiated a return to baseline CBT followed by ovulation two or more days later in the cycle. This was observed in 981 cycles (9.4%) in 939 users (14.1%)

Here is an example (Age: 30. Length of cycle: 24 days. Ovulation on Day 20):

Possible explanation: 

Researchers suggest that an initial LH surge and accompanying small progesterone rise may not always be followed by ovulation within 48 hours. Again, PCOS may be indicated here.

3) Crash after ovulation

In the crash after ovulation anomaly the final raw CBT is seen to be more than 0.2 degrees Celsius lower than the post-ovulatory-peak-averaged CBT. This was observed in 1,259 cycles (12.0%) for 1,062 users (16.0%)

Here is an example (Age: 29. Length of cycle: 38 days. Ovulation on Day 36):

Possible explanation:

Researchers suggest that progesterone may fall sharply in some women before onset of menses, and it is possible that fertility may be impaired in these cycles.

What else was observed?

Additionally, Short Luteal Phase (SLP) (d) was noted with menstruation 9 or fewer days post-ovulation – 871 cycles (8.3%); 793 users (12.0%).

SLP was also found to co-exist with the abnormal patterns [(a), (b), or (c) above in 237 cycles (2.3%); 231 users (3.5%).

Overall Implications

Could CBT monitoring offer a promising method of identifying previously undetectable causes of causes of infertility in women with ‘normal’ ovulation but who exhibit differing CBT patterns? The researchers seem to think so, particularly given the co-existence of SLP with the above patterns. They also note that ovulation generally occurs much later in each of these patterns than the textbook middle of the cycle.

And although we are not quite at the stage of AI analysis at the smartphone level, manufacturers of medical devices are already able to offer smart-algorithms. These produce reliable data analysis in real time in an instantly readable format for easy interpretation. In fact, OvuSense^TM already offers an enhanced package in OvuSense Pro^TM that produces the above charts and more just by uploading data from the sensor to a smartphone. In the OvuSense Pro^TM users can share the data with their fertility specialist for a more detailed analysis and diagnosis. For the healthcare provider this data will help determine the best course of treatment based on these ovulation patterns.

References:

1. Hurst B, et al. Atypical vaginal temperature patterns may identify subtle, not yet recognized, causes of infertility. Poster presented at the ASRM Congress. 12-16 October 2019. Philadelphia, USA. https://www.ovusense.com/uk/P345-Atypical-vaginal-temperature-patterns-may-identify-subtle-not-yet-recognised-causes-of-infertility-ASRM-2019.pdf
2. Papaioannou S, Aslam M. 2012. Ovulation assessment by vaginal temperature analysis (Ovusense Fertility Monitoring System) in comparison to oral temperature recording. Fert Steril. 98(3):S160. September 2012
3. Papaioannou S, Delkos D, Pardey J. 2014. Vaginal core body temperature assessment identifies pre-ovulatory body temperature rise and detects ovulation in advance of ultrasound folliculometry. ESHRE 30th Annual Conference. Munich, Germany. June 2014. http://www.posters2view.eu/eshre2014/data/247.pdf