Continuous glucose monitors (CGMs) have continued to prove their worth in helping people living with type 1 and type 2 diabetes better manage their condition in everyday life. Recent changes to guidelines have allowed CGM use in hospital settings for the first time, opening the door to a future where continuous glucose monitoring during hospitalization for people living with diabetes is the norm.
In this article, we examine how CGMs are currently being utilized in hospitals and what that can tell us about the future. We’ll also look at the proven benefits of using these devices in ICU and non-ICU environments, the limitations, and how innovative diabetes tech companies can turn those limitations into opportunities.
Current Trends
Continuous glucose monitors are not approved by the FDA for use in hospital settings. However, during the height of the COVID-19 pandemic, these regulations were relaxed. In a bid to help hospitals save on PPE and protect staff from unnecessary exposure, the FDA released guidance allowing for the limited use of remote patient monitoring devices for COVID-19 patients.
This change produced an unprecedented amount of data highlighting the benefits, risks, and limitations of CGM use in hospital settings. Many small-scale studies have been conducted in the past to gauge the accuracy and potential uses of CGMs in inpatient settings. But the data collected over the last two years has revealed much that these limited data-set studies did not.
While this new data is limited to ICU patients, we can use previous studies to predict how these results might change if CGMs were used in various departments throughout the hospital.
One certainty this data seems to reveal is that CGMs will be approved for use in hospital settings in the very near future.
Proven Benefits of CGM Use In Hospitals
When we look at the benefits of using CGMs in inpatient settings, we must look at ICU and non-ICU applications separately. This is due to the current differences in point of care (POC) guidelines that exist between this department and others. Even with these noted differences, there are clear advantages of CGMs over finger sticks in both settings.
Advantages Over Finger Sticks
The COVID-19 pandemic has revealed one of the most striking limitations of finger-stick POC in hospitals: the amount of time these procedures use up. This is especially true in the ICU where standard guidelines require blood sugar measurements every hour.
When you consider that each finger stick procedure requires around five minutes to complete, this time can add up fast. CGMs, on the other hand, can offer real-time instantaneous blood sugar measurements as needed. Even with the finger stick calibrations required by many models, CGM use would result in a significant reduction in time spent collecting these measurements.
More importantly, studies looking at non-ICU patients with diabetes have found that up to 45% of hypoglycemic episodes caught on CGMs are missed by finger stick POC(1). CGMs are also more than twice as likely to catch hyperglycemic events, especially overnight.
Many studies have shown a strong correlation between blood sugar levels and complications arising from procedures and hospital treatment(2). By utilizing CGMs instead of finger sticks, low and high blood sugar levels are more likely to be recorded so that action can be taken to normalize levels and avoid complications.
In ICU Settings
One of the biggest headlines to come out of the data produced by changes in hospital CGM use during the pandemic is that CGMs are no more effective at helping patients than current POC protocols. While it can be easy to take this finding to mean that CGMs are not useful in ICU settings, it is important to take a step back and consider what is really being said.
POC guidelines for patients with diabetes in the ICU are to perform finger checks once every hour. While this results in far fewer readings than what’s reported by the typical CGM (about 12 per hour), it is still far more than what non-ICU patients receive, which is only 4 to 5 finger sticks per day.
Hourly fingersticks are much less likely to miss hyper and hypoglycemic events. CGMs may add some benefit by reporting levels every 5 minutes, but their usefulness is far reduced given the typical rate of change in blood glucose over time. So while CGMs are not much more useful than POC in the ICU, they are not any less useful, either.
The second most powerful headline to come out of these studies, and the one that should be most emphasized, is that CGMs proved to be just as accurate as finger sticks(3). Meaning that either CGMs or POC can be used to produce accurate results.
In Non-ICU Patients
One striking revelation to come out of studying blood glucose trends in admitted patients is that the percentage of patients suffering from hyperglycemia (as patient-day weighted means) is nearly identical when comparing ICU and non-ICU patients(4). This similarity goes beyond the total number of hyperglycemic patients, showing nearly identical numbers between the two groups even when the data is broken down into restricted ranges, as illustrated below.
| Glucose Level | ICU Patient-Days, % | Non-ICU Patient-Days, % |
| Below 180 | 32.2% | 32.0% |
| 180 to 250 | 23.1% | 23.0% |
| 250 to 300 | 9.8% | 9.9% |
| 300 to 350 | 4.0% | 4.1% |
| 350 to 400 | 1.6% | 1.6% |
| Over 400 | 0.6% | 0.6% |
What this tells us is that non-ICU patients are just as likely to suffer from complication-inducing hyperglycemia as ICU patients. Yet, standard POC calls for hourly blood sugar checks for the latter and only mealtime and bedtime checks for the former.
Of course, many non-ICU departments are not staffed to allow for such frequent patient checks. And this is where CGMs could be highly beneficial. Compared to current POC fingerstick standards, a CGM is far more likely to catch both hyper and hypoglycemic episodes. They can also be set up to alarm at the nurses’ station if such an episode occurs. This means no extra time would be required for blood sugar checks when no problems exist.
With CGM monitoring, doctors are also able to get a more complete picture of the patient’s response to insulin and medications known to affect blood sugar. This allows them to make educated adjustments to the patient’s treatment to assure more time is spent within a normal glucose range.
Limitations of CGM Use in Hospitals
There are clear benefits to using CGMs in hospital settings, especially in non-ICU departments where current POC offers limited glucose monitoring. But these tools are not without their limitations, many of which are specific to hospital settings.
Staff Training
As with any novel medical technology, training would be necessary to get staff familiar with using CGMs as the main mode of glucose monitoring. Learning how to interpret and react to the constant influx of glucose data is paramount. Especially considering the potential for treatment overload that could come with such a vast amount of data.
Medical staff would also need to be trained to apply the sensors. The best current candidates for hospital use are CGMs made up of a sensor that’s inserted below the skin with a handheld device and an attachable transmitter. Learning how to properly place and deploy these sensors is important to attaining accurate results.
Additionally, staff would need training on using the connected receiver and applicable mobile applications.
Compatibility with EHR Systems
One current limitation that could quickly reduce the training burden if overcome is integrating CGM tech with Electronic Health Record (EHR) systems. Because CGMs are not approved for hospital use, this integration is nonexistent. In COVD ICUs, medical staff is forced to use such rudimentary methods as taping CGM receivers to windows in order to safely read and react to CGM data.
Integrated EHR tech would allow staff to reference glucose information from their station using the same screen they use to reference all other vitals information. It would also allow for system-wide alerts if dangerously high or low blood sugars were predicted. Treating doctors could access glucose trends remotely from their offices and establish treatment adjustments with this and other parameters in mind.
Without a doubt, EHR compatibility will weigh heavily on which CGM systems hospitals choose when making the switch after this technology is inevitably approved for hospital use.
Substance and Treatment Interference
Current CGM systems have shown a potential for inaccurate results when certain medications and treatment options are used. For instance, it is well established that induced or resultant hypothermia can cause highly inaccurate CGM readings. Additionally, the use of bypass frequently results in CGM failure, requiring other means of glucose monitoring during certain cardiac surgeries.
A long list of drugs, including lisinopril, albuterol, acetaminophen, atenolol, dopamine, mannitol, and heparin, have been found to skew subcutaneous glucose readings(5,6). We are seeing more studies into these types of interactions since COVID, but more research will be needed to understand how CGM use might be affected by hospital-specific drugs and dosages.
Sensors that are less sensitive to commonly used hospital drugs will be preferred over those that are highly inaccurate in the typical hospital setting.
Condition Interference
Beyond treatment interference, current CGMs have also shown the potential for inaccuracy when used in patients with certain medical conditions. Most notably, anemia can skew CGM glucose readings and make them unusable. Other conditions that have shown some potential for interference include hypotension and hypoxia(6).
The current list of medical conditions that interfere with CGM readings is based almost entirely on conditions likely to be seen in ICU settings. Further study is needed to determine what other conditions commonly seen in hospitals, but less common in ICUs, may impact CGM accuracy.
Once this is known, it will be highly advantageous and important for medical tech companies to develop sensors that can overcome these limitations.
Ongoing CGM Limitations
Beyond hospital-specific CGM limitations is a long list of limitations already well-established with the normal use of these devices.
The high cost of CGM sensors and transmitters may keep some hospitals from being able to make the switch to this technology. Lag time, which is seen to some degree in all subcutaneous sensors, is also problematic when accurate, up-to-the-minute data is required. Overall accuracy also continues to be a problem with available CGMs when glucose levels fall below or above a certain point. And the need for calibrations to maintain accuracy or start sensors is perceived as a limitation by many since it requires POC tools while asking hospitals to get away from current POC protocols.
Continuing to improve on these well-established and ongoing CGM limitations will be vital for implementing CGM use in hospitals.
Limitations as Opportunities for Future Development
While it can be easy to look at the above list as inhibitors to success in the CGM market, aspiring medical technology companies should really see these limitations as opportunities. Incompatibility, substance and condition interference, and ongoing use limitations all present an opportunity for the next generation of CGMs to improve on those currently available. And to do so in a way that would make these CGMs far more valuable to both inpatient use and outpatient use applications.
There is no doubt that FDA approval of widespread CGM use in hospitals is on the horizon, with many experts agreeing it will happen within a year. New products that can overcome current limitations will be in high demand as soon as this regulation shift occurs. For innovative, performance-minded medical technology and SaMD companies, this new market promises to be a gold mine of opportunity.