End-to-End Sepsis Solution Incorporating Expert Telemedicine Consultation

Abstract

Introduction:

Early detection and optimal resuscitation of critically ill sepsis patients may improve sepsis care delivery. The objective was to assess the feasibility of developing and implementing an end-to-end sepsis solution including early detection, monitoring, and teleconsultation.

Methods:

Prospective implementation of an end-to-end sepsis solution for potential sepsis patients presenting to a community hospital emergency department (ED) between 11 AM and 5 PM, Monday to Friday, during a 40-day period in 2019. Qualifying patients were compared with patients presenting at other times during the pilot screening period and to historic controls.

Results:

During the initial period, 203 patients met the screening criteria for potential sepsis; 77 patients (37.9%) had a primary diagnosis of sepsis, present on admission. Mean age was 60 ± 20 years; 50.7% were female; and 24 patients (11.8%) were primary sepsis, SEP-1 bundle eligible. Eighty of 203 (39.4%) had an initial lactate performed, mean, 2.7 ± 1.7 mmol/L. For the 24 primary sepsis, SEP-1 bundle eligible patients, 100% received antibiotics and intravenous fluid. Thirteen consults were performed on 12 patients; mean time from consult decision to beam in to the telemedicine robot was 7.3 ± 5.5 min; mean time from beam in to robot connection with the expert was 23.6 ± 13.2 s; mean consultation call time was 6.3 ± 4.3 min.

Conclusions:

In a convenience sample of patients with potential sepsis presenting to a community hospital ED, an end-to-end sepsis solution using early detection, tracking, and consultation was feasible and has the potential to improve sepsis detection and treatment.

Introduction

Sepsis, defined as a dysregulated physiologic response to an infection, can cause acute organ dysfunction (AOD) and lead to disability and death. Between 1,000,000 and 2,000,000 cases occur annually in the United States and the mortality rate ranges from 15% to 25%.^1
–3 Protocolized clinical trials and focused database research have emphasized the importance of early detection, timely initiation of treatment, and delivery of a core bundle of care in improvement of outcomes,^4

–7 and these principles are emphasized by the Surviving Sepsis Campaign. Hospitals that adhere to this approach have demonstrated a lower mortality rate.⁸

Controversy exists about the optimal screening strategy to use when a potential sepsis patient presents to the emergency department (ED). The original 1992 International Sepsis definitions, promulgated in 1992, used the Systemic Inflammatory Response Syndrome (SIRS) criteria as screening variables to begin capturing sepsis patients.⁹ After these criteria were initially codified, researchers investigated their utility for capturing undifferentiated patients with potential infection and categorizing recently diagnosed patients.¹⁰ An evolving sense that the SIRS criteria were neither sensitive nor specific enough to effectively capture sepsis patients at the proximal phase of medical care, led to the search for other screening strategies.¹¹ The Third International Sepsis Definitions proposed replacing SIRS with a quick Sequential Organ Failure Assessment (qSOFA) score and presented data using the qSOFA criteria that could be used at the bedside to rapidly identify patients at increased risk of death or prolonged intensive care unit stay during their admission.¹²

Subsequent research has suggested that qSOFA is less sensitive but more specific than SIRS criteria for the initial detection of sepsis patients and demonstrated that a combination of the two yields better results.^13

–16 The potential exists to deploy human or automated resources to screen for potential sepsis patients at the proximal phase of presentation. These screening and capture tools can also be linked to expedited practitioner assessment, and initiation of diagnosis and treatment, including appropriate antibiotics.^17
–19

Another complicating factor in optimizing sepsis care is that sepsis patients present to every hospital in the United States and these hospitals have highly variable resources, staffing, and specialty care capabilities. Hospitals that manage a larger volume of sepsis patients have better outcomes across increasing levels of disease severity. Telemedicine solutions have the potential to enable remote physician resources to provide monitoring, bundle compliance assistance, expert consultation, and act as practitioner extenders in understaffed settings.^{17,20

–24}

In this feasibility study, we explored whether a nurse and physician monitor could capture potential sepsis patients real time after ED triage, identify AODs as data became available, detect “time zero” for sepsis, and identify the most critically ill subset of sepsis patients who might benefit from expert teleconsultation. Primary outcomes include the percentage of patients with a primary sepsis, present on admission (POA); time to bundle measures including initial lactate, intravenous fluid (IVF), antibiotics, and repeat lactate; and time from consult decision to connection, length of telemedicine sepsis consultation, and quality of telemedicine connection (video health).

Methods

STUDY DESIGN

We performed a prospective, real-time study of patients meeting screening criteria for potential sepsis with AOD. This study was approved by the Thomas Jefferson University's Institutional Review Board and was granted a waiver of informed consent.

STUDY SETTING AND TIME PERIOD

The study was performed Monday to Friday, 11 AM to 5 PM, January 28, 2019, to March 22, 2019, in the ED of a community hospital affiliated with an academic teaching hospital. In 2019, the ED saw 39,874 patients, and during the study dates, 5,808 patients were triaged.

STUDY POPULATION

Adult patients (≥18 years old) presenting to the ED were screened based on two categories of variables: (1) The presence of three or more clinical screening criteria associated with sepsis patients who have AOD; and (2) documented chief complaint consistent with a presentation for sepsis. The clinical screening criteria consisted of a combination of SIRS and qSOFA criteria: temperature <36°C (96.8°F) or >38°C (100.4°F); heart rate >90 beats per minute; respiratory rate >20 breaths per minute (or PCO₂ <32 mmHg); white blood cell count <4,000/mm³, >12,000/mm³, or >10% immature cells; systolic blood pressure ≤100 mmHg; and altered mentation (Glasgow Coma Score <15). Chief complaints consistent with infection include the following: fever, chills, shortness of breath, dyspnea on exertion, chest pain, and abdominal pain (Supplementary Table S1).

For comparative data, we obtained a historic control set of patients with a discharge diagnosis of sepsis, POA who were discharged between September 1, 2018, and November 30, 2018, and a contemporaneous set of patients with sepsis POA discharged during the study period, January 28, 2019, to March 22, 2019. The classification of these patients as sepsis POA was reviewed and confirmed or reclassified by one of the authors (D.F.G.).

DATA COLLECTION

Patients triaged during study hours were monitored in the electronic medical record (EMR) (EPIC, Verona, WI) and entered into Smart Notes (InTouch Health, Santa Barbara, CA) if they had ≥3 screening criteria over the course of their ED stay. Monitoring was performed by one of the teleconsultation sepsis experts (D.F.G.), a research nurse, and a technician from InTouch Health (J.C.), who were either on site monitoring the EMR from a nonclinical space or monitoring the EMR remotely. A call schedule was made for the teleconsultation experts, who had no other clinical duties during their teleconsultation shifts.

Data points collected during the ED stay included the following: cumulative screening criteria, age, race/ethnicity, time of arrival, chief complaint, vital signs, laboratory values including white blood cell count, lactate, platelets, total bilirubin, creatinine, and International Normalised Ratio (INR); time, type, and quantity of fluid administered; time and name of antibiotics administered; and admission or discharge. Patients who met the criteria for enrollment in the CMS SEP-1 sepsis performance bundle were assigned a “time zero” in Smart Notes, defined as the time when the patient met the screening criteria, a documented or suspected infection, and had evidence of AOD.

Patients entered into Smart Notes were classified into the following categories: sepsis primary, bundle eligible; sepsis primary, not bundle eligible; sepsis not primary or no sepsis; sepsis primary, not diagnosed during ED stay; sepsis primary, limitations of care established during ED stay.

Sepsis primary patients were eligible for a telemedicine sepsis consultation if: the initial lactate level was >4 mmol/L; the patient had persistent hypotension (systolic blood pressure [SBP] <90 mmHg or mean arterial pressure [MAP] <65 mmHg) after 30 cc/kg fluid bolus; ≥3 AODs attributable to sepsis.

We calculated time zero; time from triage to first lactate result; time from time zero to initiation of fluids and antibiotics; and time from first lactate result to repeat lactate draw. For patients receiving a telemedicine consultation, we calculated time from the decision to perform a consult to beam in to the telerobot; time from beam in to connection; length of the consultation; video health (rated 0–100); and notes specific to the call.

STATISTICAL ANALYSIS

For descriptive analysis, continuous data are expressed as means (±standard deviations) and differences were tested using Student's t-test. Categorical variables are presented as percentages and analyzed using the chi-squared test or Fisher's exact method. All analyses were completed using statistical software embedded in Excel (MicroSoft^®, Redmond, WA).

Results

DEMOGRAPHICS AND SEPSIS CATEGORIES

During the 40-day telesepsis evaluation period, 203 patients met the screening criteria for potential sepsis; mean age was 60 ± 20 years; 50.7% were female; 77 patients (37.9%) had a primary diagnosis of sepsis, POA. Fifty (24.6%) were sepsis primary, not bundle eligible; 24 patients (11.8%) were sepsis primary, SEP-1 bundle eligible (Table 1).

Table 1.

Demographics and Sepsis Categories

	SEPSIS POA HISTORIC CONTROLS	TELESEPSIS PILOT	SEPSIS POA DURING TELESEPSIS PILOT PERIOD	p
Number	122	203	112
Age
Mean, years	65.2	60	66	0.012
SD, years	19.2	20	19.3	0.012
Sex
Female, n (%)	55 (45.1)	103 (50.7)	55 (49)	0.98
Male, n (%)	67 (54.9)	100 (49.3)	57 (51)	0.98
Category
Sepsis primary, bundle eligible, n (%)	82 (67.2)	24 (11.8)	72 (63.2)	<0.001
Sepsis primary, not bundle eligible, n (%)	25 (20.5)	50 (24.6)	34 (29.8)
Sepsis not primary or no sepsis, n (%)	14 (11.5)	126 (62.1)	8 (7)
Sepsis primary, care limitations during bundle period, n (%)	1 (<1)	2 (<1)	0 (0)
Sepsis primary, not diagnosed in ED, n (%)	0 (0)	1 (<1)	0 (0)

ED, emergency department; POA, present on admission; SD, standard deviation.

During the 3-month historic control period, 122 patients were coded as sepsis POA; mean age was 65.2 ± 19.2 years; 45.1% were female; after review, 108 patients (88.5%) had a final diagnosis of sepsis primary, POA, and 82 (67.2%) were sepsis primary, POA, SEP-1 bundle eligible (Table 1).

During the 2-month concurrent data collection period, 114 patients were coded as sepsis POA; after review, 2 patients were excluded because they were admitted before the study period (1 patient) or were directly admitted to the hospital, bypassing the ED (1 patient), leaving a final cohort of 112 patients; mean age was 66 ± 19.3 years; 49% were female; after review, 104 (91.2%) had a final diagnosis of sepsis primary, POA, and 72 (63.2%) were sepsis primary, POA, SEP-1 bundle eligible (Table 1).

SEPSIS CARE METRICS

Telesepsis pilot

Eighty of 203 (39.4%) telesepsis pilot potential sepsis patients had an initial lactate performed, the mean value was 2.7 ± 1.7 mmol/L, and 43 (53.8%) of these were >2 mmol/L. For the 24 patients who were sepsis primary, POA, SEP-1 bundle eligible, 100% received antibiotics and IVF (Table 2).

Table 2.

Sepsis Care Metrics

	SEPSIS POA HISTORIC CONTROLS	TELESEPSIS PILOT POA PATIENTS	SEPSIS POA DURING TELESEPSIS PILOT PERIOD	p
Number	122	203	112
Initial lactate
n (%)	111 (90.1)	80 (39.4)	98 (87.5)	<0.001
Mean, mmol/L	2.9	2.7	3.3	0.18
SD, mmol/L	2.2	1.7	2.6	0.18
Lactate >2 mmol/L, n (%)	68 (60.4)	43 (53.8)	62 (63.3)	0.76
Lactate >3 mmol/L, n (%)	31 (27.9)	25 (31.3)	36 (36.7)
Lactate >4 mmol/L, n (%)	17 (15.3)	14 (17.5)	24 (24.5)
Repeat lactate
n (%)	74 (66.7)	39 (48.8)	54 (55.1)	0.417
Mean, mmol/L	2.6	2.7	3.3
SD, mmol/L	2.8	1.9	2.8	0.303
Lactate >2 mmol/L, n (%)	31 (41.9)	19 (48.7)	29 (53.7)	0.96
Lactate >3 mmol/L, n (%)	13 (17.6)	9 (23.1)	17 (31.5)
Lactate >4 mmol/L, n (%)	10 (13.5)	7 (18)	12 (22.2)
First antibiotic
n (%)	115 (94.3)	75 (37)	102 (91.1)	<0.001
First antibiotic, if any sepsis primary, n (%)	107 (99.1)	56 (72.7)	100 (96.2)	0.46
First antibiotic, sepsis primary, bundle eligible, n (%)	82 (100)	24 (100)	67 (100)	1
Second antibiotic, if first
n (%)	88 (76.5)	37 (49.3)	94 (92.2)	0.039
First antibiotic, if any sepsis primary, n (%)	76 (71)	37 (66.1)	80 (80)	0.732
First antibiotic, sepsis primary, bundle eligible, n (%)	70 (85.4)	20 (83.3)	62 (92.5)	0.928
Fluid
Any fluid, n (%)	119 (97.5)	126 (62.1)	101 (90.2)	0.017
Mean, cc	1,805	1,363	1,401	<0.001
SD, cc	911	820	722	<0.001
Sepsis primary, bundle eligible, any fluid, n (%)	82 (100)	24 (100)	67 (100)	1
Mean, cc	1,943	1,839	1,456	0.002
SD, cc	890	820	741	0.002
Sepsis primary, not bundle eligible, any fluid, n (%)	24 (96)	33 (69)	30 (81.1)	0.512
Mean, cc	1,479	1,464	1,278	0.565
SD, cc	905	851	624	0.565

Sepsis POA historic controls

One hundred eleven of 122 (90.1%) sepsis POA historic control patients had an initial lactate performed, the mean value was 2.9 ± 2.2 mmol/L, and 68 (60.4%) of these were >2 mmol/L. For the 82 patients who were sepsis primary, POA, SEP-1 bundle eligible, 100% received antibiotics and IVF (Table 2).

Sepsis POA during telesepsis pilot period patients

Ninety-eight of 112 (87.5%) sepsis POA during telesepsis pilot period patients had an initial lactate performed, the mean value was 3.3 ± 2.6 mmol/L, and 62 (63.3%) of these were >2 mmol/L. For the 67 patients who were sepsis primary, SEP-1 bundle eligible, 100% received antibiotics and IVF (Table 2).

TIME TO SEP-1 REPORTING METRICS

Times to sepsis quality metrics were statistically different, but clinically similar between historic control, pilot period screening, and pilot period sepsis POA patients (Table 3).

Table 3.

Time to SEP-1 Reporting Metrics

	METRIC
	TIME TO FIRST LACTATE	TIME TO IVF	TIME TO ANTIBIOTICS	TIME TO REPEAT LACTATE
Group
Historic, sepsis POA, bundle eligible, mean	01:32	0:22	1:11	4:00
SD	0.047	0.042	0.067	0.093
Telesepsis pilot, bundle eligible, mean	1:06	0:33	1:28	4:58
SD	0.031	0.035	0.052	0.150
Telesepsis pilot period, sepsis POA, bundle eligible, mean	1:28	0:23	0:55	3:52
SD	0.054	0.031	0.044	0.081
p-Value, three groups	<0.001	<0.001	<0.001	<0.001
p-Value, historic POA vs. pilot period POA	<0.001	0.049	<0.001	<0.001
Time calculation	Arrival to first lactate result	Sepsis time zero to first IVF	Sepsis time zero to first antibiotics	First lactate result to second lactate draw

IVF, intravenous fluid.

CONSULT CALL SUMMARY, DETAILS, AND NOTES

During the telesepsis pilot period, a total of 19 calls were made, 3 were test calls, and 1 call was interrupted when the robot shut down. Thirteen consults were performed on 12 patients; the average time from consult decision to beam in to the telemedicine robot was 7.3 ± 5.5 min; the average time from beam in to connection of the robot with the expert consultant was 23.6 ± 13.2 s; the average consultation call time was 6.3 ± 4.3 min. Median video health for the calls was 100 and the range was 10–100. Comments entered in regard to the calls included “Follow up consult; Dr. A supervised ultrasound,” “Robot shut off; initial reconnect failed,” and “Attending with another patient; consultant spoke with patient directly” (Tables 4 and 5).

Table 4.

Consult Call Detail Report

PATIENT NUMBER	CONTACT RESIDENT/ATTENDING TIME	BEAM IN TIME	TIME FROM CONSULT DECISION TO BEAM IN, MIN	BEAM IN TO CONNECT, S	VIDEO HEALTH	DURATION	EXPERT CONSULTANT ON-CALL
1	1:03 PM	1:04 PM	6	17	100	00:05:58	A
1	2:32 PM	2:33 PM	1	13	100	00:06:45	A
2	12:44 PM	12:50 PM	8	20	100	00:06:50	A
TEST	xxxxx	xxxxx	xxxxx	xxxxx	xxxxx	00:00:00	A
TEST	xxxxx	xxxxx	xxxxx	xxxxx	100	00:00:25	B
TEST	1:44 PM	1:52 PM	8	32	100	00:00:19	B
3	xxxxx	2:33 PM	xxxxx	xxxx	100	00:14:28	B
4	12:35 PM	12:43 PM	8	24	100	00:07:59	A
5	xxxxx	xxxxx	xxxxx	xxxx	43	00:09:02	B
5	xxxxx	xxxxx	xxxxx	28	xxxxx	00:00:00	B
5	4:35 PM	4:42 PM	7	xxxxx	80	00:09:43	B
6	2:14 PM	2:35 PM	21	60	100	00:11:01	A
7	1:21 PM	1:22 PM	1	16	100	00:06:46	A
8	4:51 PM	4:54 PM	3	13	100	00:05:17	A
9	3:28 PM	3:34 PM	6	13	100	00:08:14	A
10	3:35 PM	3:40 PM	5	17	100	00:08:27	A
11	3:27 PM	3:40 PM	13	29	10	00:00:20	A
11	xxxxx	xxxxx	xxxxx	xxxxx	100	00:10:00	A
12	1:19 PM	1:24 PM	7	18	100	00:08:01	C

STATISTIC	CONSULT DECISION TO BEAM IN TIME	BEAM IN TO CONNECT	VIDEO HEALTH	DURATION
Mean	7.33	23.58	89.5	06:19
SD	5.45	13.24	25.8	04:22
Median	7	19	100	07:24
Minimum	1	13	10	00:00
Maximum	21	60	100	14:28
Units	Min	S	0–100 Scale	Min:S

xxxxx, not applicable.

Table 5.

Notes and Summary

NOTES
Initial consult on patient
Follow-up consult; Dr. A supervised ultrasound
Incorrect number listed for resident
Test
Test
Test: IT fixing Dr. B's computer
Follow up to perform consult
Resident initially with another patient
Patient receiving treatment; resident in room
Robot shut off; initial reconnect failed
Robot shut off; had to reconnect to complete consult
Patient consented to record consultation
MD with another patient; spouse to consent
Patient at test; found resident waiting at MD station
Shift change for residents, consult during handoff
Attending with another patient, consultant spoke with patient directly
Spouse completing consent forms
Consult completed, spouse present
Needed time to inform patient of consult

SUMMARY
Total calls: 19
Consults: 13
Individual patients: 12
Consults by consultant:
A: 10 (9)
B: 2 (2)
C: 1 (1)
Test calls: 3
Robot shut down: 1

SENSITIVITY AND SPECIFICITY OF TELESEPSIS PILOT SCREENING FOR CAPTURE OF SEPSIS POA PATIENTS

During the telesepsis pilot, 203 patients screened positive and 77 of these patients were classified as sepsis primary, POA (specific 37.9%), 26 of these patients had AOD meeting criteria for SEP-1 bundle compliance, and 24 of these patients were adjudicated as having sepsis primary, POA, and be SEP-1 bundle eligible (specificity 11.8%). Two patients (7.7%) were not bundle eligible because of limitations on level of care established during their ED stay.

When the 67 pilot time frame, sepsis primary, POA, bundle eligible patients captured by International Classification of Diseases-10 discharge codes for severe sepsis or septic shock were limited to the time frame of the telesepsis screening pilot (M-F, 11A-5P), there was a total of 18 patients, 100% of whom were captured by the telesepsis pilot screening protocol (sensitivity 100%). The additional nine sepsis primary, POA, bundle eligible patients captured by the telesepsis screening protocol all had infection, AOD, and were treated for severe sepsis in the ED, receiving lactate measurements, antibiotics, and IVF. One hundred percent of these patients had discharge codes for infection (e.g., pneumonia) and codes for organ dysfunction (e.g., acute kidney injury), but no codes for sepsis, severe sepsis, or septic shock.

Discussion

In this study, we found that we were able to implement an end-to-end sepsis solution, using EMR screening, an on-site nurse navigator, telemedicine expert consultation, and real-time data collection, to capture, track, monitor, and assist in the management of sepsis patients. The patients with a primary diagnosis of sepsis who were SEP-1 bundle eligible during the evaluation period were similar demographically to a cohort of patients with a primary sepsis diagnosis POA the previous year in the same ED and to all patients discharged from the hospital during the study period with a primary sepsis diagnosis POA and admitted through the ED. Our end-to-end sepsis solution was more sensitive than discharge codes for capturing patients treated for sepsis in the ED and during their in-patient stay.

Only 17 of the 26 patients, SEP-1 eligible patients, were captured by discharge codes, consistent with other work showing lack of sensitivity of coding for capture of sepsis cases.²⁵ This has implications for proper treatment, documentation of burden of sepsis cases, and benchmarking quality of care.

We demonstrated that a manual sepsis screening tool, combining SIRS and qSOFA criteria with a chief complaint concerning an infectious diagnosis, was highly sensitive to capture patients admitted to the hospital with sepsis. As expected, this sensitive screening technique had low specificity for sepsis (77/203; 37.9%) and even lower for sepsis with AOD, SEP-1 bundle eligible (24/203; 11.8%). In a next-generation tool, this approach could be automated in the EMR and linked to best practice alerts or clinical advisories. The combination of screening criteria chosen has the potential to overcome limitations of both SIRS and qSOFA as isolated approaches.

In this work, we did not intervene clinically except when patients met the criteria for a telesepsis consultation. Patients meeting sepsis screening criteria were managed by the clinical staff as they judged appropriately and the study had no influence on the length of time a patient waited in the waiting room or time to intravenous access, decision to screen the patients with initial or repeat lactate measurements, or decisions to start IVFs, antibiotics, or admit the patient to the hospital. The exception to this was when patients had a telesepsis consultation, none of which occurred before when an initial lactate sample was sent and the start of resuscitation, fluids, and antibiotics.

We found that sepsis care metrics varied between the three groups. The vast majority of primary sepsis, POA historic control, and telesepsis evaluation period patients had an initial lactate performed in the ED (90.1% vs. 87.5%, respectively), whereas patients meeting the initial screening criteria during the telesepsis pilot had a much lower level of initial lactate screening (39.4%; p < 0.001). This finding is as we predicted since the screening criteria signal potential sepsis not confirmed sepsis. In a fully operationalized sepsis screening, monitoring, and treatment algorithm, patients who met the screening criteria would be immediately assessed by a clinician and classified as “obviously not sepsis,” “concern for sepsis,” or “definitive sepsis.” Patients with a concern for sepsis can be prioritized for laboratory and imaging studies, including initial lactate measurement, to assess for the presence of infection and AOD, not diagnosable on the initial screening variables and clinical assessment.

Time to SEP-1 quality reporting metrics was statistically different but clinically similar between the groups. For example, the time from arrival in triage to first lactate result ranged from 1:06 (h:min) in the telesepsis pilot, bundle eligible group to 1:32 (h:min) in the sepsis primary, POA, bundle eligible historic controls. Similar results were seen for time to first IVF, first antibiotics, and repeat lactate draw. These mean times fall within the recommended time frames codified by SEP-1 including <3 h to first lactate result and start of appropriate antibiotics and <6 h to repeat lactate sample being obtained if the first lactate was >2 mmol/L.²⁶ However, earlier identification of these patients by our screening methodology provides the opportunity to notify clinical staff of the presence of a possible sepsis patient and expedite their care.²⁷

Twelve of the 24 telesepsis pilot, bundle eligible patients (50%) met the criteria for a telesepsis consult and consults were performed on all of these patients, including 2 discrete consults on 1 patient. The mean length of a consultation was 6:19 ± 4:22 min:sec and time from decision to perform a consultation to beaming in to the robot, time from beaming in to the robot to connection with the clinical staff, and video health were all acceptable to the clinical staff involved. These consultations led to additional imaging, antibiotics, diagnostic tests, supervision of procedures including ultrasounds and central venous catheter placement, discussion of the care plan with patients and family, and discussion of goals of care with patients and family.

Implementing and automating the steps in this project—detection, monitoring, consultation, and data collection—and linking them to communications to expedite care have the potential to capture a higher percentage of sepsis patients in a timely manner, improve times to SEP-1 metrics, and improve clinical outcomes including hospital length of stay, in-hospital mortality rate, and the percentage of patients discharged to home.^17,18 This may also allow state-of-the-art sepsis care to be delivered at institutions with a lower sepsis case volume, with the potential to improve outcomes to those seen at high-volume sepsis centers.²⁸

LIMITATIONS

The telesepsis pilot group had a relatively small number of patients and precluded the meaningful comparison of certain metrics between groups. However, the groups were well-balanced on age, sex, and sepsis metrics. Second, the telesepsis pilot group was conducted during a convenience period in the middle of the day during weekdays and this raises the possibility that the composition of the pilot group was different from the patients as a whole. The fact that metrics including antibiotic administration, fluid administration, and fluid volume were not statistically significantly different between the groups when each group was limited to sepsis primary, bundle eligible patients suggests that the groups were similar. Third, availability of study personnel on site to discuss potential consultations with bedside clinical staff may have increased compliance with recommended consultations, and further studies would need to be performed to assess the effectiveness of this management strategy.

Finally, similarly, the fact that the consultant sepsis experts were well known to the bedside clinical staff in the ED may have improved consult compliance and further studies would be needed to assess the effectiveness of consultations when performed in a larger health system.

Conclusions

In this pilot study, we demonstrated the feasibility of implementing an end-to-end sepsis solution incorporating early detection, monitoring, telesepsis consultation, and data collection. Teleconsultation has the potential to improve sepsis care, increase SEP-1 compliance, augment bedside staff, and supervise resident procedures.

Footnotes

Disclosure Statement

No competing financial interests exist.

Funding Information

This research was supported by funding from a joint venture between Jefferson Strategic Ventures and InTouch, Inc.

Supplementary Material

Supplementary Table S1

References

Singer

, Deutschman

, Seymour

, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA, 2016; 315(8):801–810; doi: 10.1001/jama.2016.0287

Gaieski

, Edwards

, Kallan

, et al. Benchmarking the incidence and mortality of severe sepsis in the United States. Crit Care Med, 2013; 41(5):1167–1174; doi: 10.1097/CCM.0b013e31827c09f8

Kaukonen

K-M

, Bailey

, Suzuki

, et al. Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000–2012. JAMA, 2014; 311(13):1308–1316; doi: 10.1001/jama.2014.2637

Rivers

, Nguyen

, Havstad

, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med, 2001; 345(19):1368–1377; doi: 10.1056/NEJMoa010307;

The ProCESS Investigators. A randomized trial of protocol-based care for early septic shock. N Engl J Med, 2014; 370(18):1683–1693; doi: 10.1056/NEJMoa1401602

The ARISE Investigators and the ANZICS Clinical Trials Group. Goal-directed resuscitation for patients with early septic shock. N Engl J Med, 2014; 371(16):1496–1506; doi: 10.1056/NEJMoa1404380

Mouncey

, Osborn

, Power

, et al. Trial of early, goal-directed resuscitation for septic shock. N Engl J Med, 2015; 372(14):1301–1311; doi: 10.1056/NEJMoa1500896

Levy

, Rhodes

, Phillips

, et al. Surviving Sepsis Campaign: Association between performance metrics and outcomes in a 7.5-year study. Crit Care Med, 2015; 43:2–12.

Bone

, Balk

, Cerra

, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest, 1992; 101:1644–1655.

10.

Rangel-Frausto

, Pittet

, Costigan

, et al. The natural history of the systemic inflammatory response syndrome (SIRS), A prospective study. JAMA, 1995; 273:1117–1123.

11.

Kaukonen

K-M

, Bailey

, Pilcher

, et al. Systemic inflammatory response criteria in defining severe sepsis. N Engl J Med, 2015; 372:1629–1638.

12.

Seymour

, Liu

, Iwashyna

, et al. Assessment of clinical criteria for sepsis: For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA, 2016; 315(8):762–774; doi: 10.1001/jama.2016.0288

13.

Churpek

, Snyder

, Han

, et al. Quick sepsis-related organ failure assessment, systemic inflammatory response syndrome, and early warning scores for detecting clinical deterioration in infected patients outside the ICU. Am J Respir Crit Care Med, 2017; 195(7):906–911; doi: 10.1164/rccm.201604-0854OC

14.

Freund

, Lemachatti

, Krastinova

, et al. Prognostic accuracy of sepsis-3 criteria for in-hospital mortality among patients with suspected infection presenting to the emergency department. JAMA, 2017; 317(3):301–308; doi: 10.1001/jama.2016.20329

15.

Kim

, Suh

, Kim

, et al. Quick Sepsis-related Organ Failure Assessment score is not sensitive enough to predict 28-day mortality in emergency department patients with sepsis: A retrospective study. Clin Exp Emerg Med, 2019; 6(1):77–83.

16.

Perman

, Mikkelsen

, Goyal

, et al. The sensitivity of qSOFA calculated at triage and during emergency department treatment to rapidly identify sepsis patients. Sci Rep, 2020; 10:20539.

17.

Mohr

, Campbell

, Swanson

, et al. Provider-to-provider telemedicine improves adherence to sepsis bundle care in community emergency departments. J Telemed Telecare, 2021; 27:518–526.

18.

Machado

, Wilson

, Elliott

, et al. Impact of a telemedicine eICU cart on sepsis management in a community hospital emergency department. J Telemed Telecare, 2018; 24:202–208.

19.

Tande

, Berbari

, Ramar

, et al. Association of a remotely offered infectious disease eConsult with improved clinical outcomes. Open Forum Infect Dis, 2020; 17:ofaa003.

20.

Badawi

, Hassan

. Telemedicine and the patient with sepsis. Crit Care Clin, 2015; 31:291–294.

21.

Deisz

, Rademacher

, Gilger

, et al. Additional telemedicine rounds as a successful performance-improvement strategy for sepsis management: Oberservational multicenter study. J Med Internet Res, 2019; 21:e11161.

22.

Davis

, Faruki

, Breyer

, et al. The case for virtual sepsis surveillance and intervention. Telemed J E Health, 2022; 28:102–106.

23.

Marx

, Greiner

, Juhra

, et al. An innovative telemedical network to improve infectious disease management in critically ill patients and outpaitnets: A stepped-wedge, cluster randomized control trial (TELnet@NRW). J Med Internet Res, 2022; 24(3):e34098; doi: 10.2196/34098

24.

Rincon

, Bourke

, Seiver

. Standardizing sepsis screening and management via a tele-ICU program improves pateint care. Telemed J E Health, 2011; 17:560–564.

25.

Whittaker

S-A

, Mikkelsen

, Gaieski

, et al. Severe sepsis cohorts derived from claims-based strategies appear to be biased towards a more critically ill patient population. Crit Care Med, 2013; 41:945–953.

26.

SEP-1 bundles. Available from: https://cmit.cms.gov/CMIT_public/ViewMeasure?MeasureId=00678

27.

Gaieski

, Pines

, Band

, et al. Impact of time to antibiotics on survival in patients with severe sepsis or septic shock in whom early goal-directed therapy was initiated in the emergency department. Crit Care Med, 2010; 38(4):1045–1053; doi: 10.1097/CCM.0b013e3181cc4824

28.

Gaieski

, Edwards

, Kallan

, et al. The relationship between hospital volume and mortality in severe sepsis. Am J Respir Crit Care Med, 2014; 190:665–674.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.01 MB

0.00 MB