Exploring the role of point-of-care ultrasound for people with HIV in a resource affluent setting: A prospective observational study

Abstract

Background

In low resource settings point of care ultrasound (POCUS) has proven value for the detection of opportunistic diseases in HIV, especially tuberculosis. Few studies have explored POCUS in people with HIV and potential opportunistic infections in resource affluent settings.

Methods

We performed a prospective observational study in the Netherlands. Outpatients newly diagnosed with HIV and a CD4 T-cell count below 350 cells/mm³ and inpatients with HIV on the infectious diseases ward were included. POCUS of chest and abdomen were performed to detect opportunistic diseases and patients were followed for 1 year to register the presence or absence of opportunistic diseases as detected during routine care. Primary outcome was the number of HIV-related and unrelated conditions detected by POCUS and interobserver variation (ClinicalTrials.gov registration NCT04246983).

Results

We included 34 patients (79% males, median CD4 T-cell count 115/mm³). Observers had 97.5% agreement (Cohen’s kappa for interobserver reliability 0.75). POCUS examination indicated at least one new opportunistic disease in 11 patients. In all these patients additional investigations confirmed opportunistic disease, predominantly Pneumocystis jirovecii pneumonia, bacterial pneumonia, Mycobacterium avium complex infection and lymphoma. In four patients an opportunistic disease was diagnosed in foci inaccessible for POCUS and a normal POCUS correctly excluded opportunistic diseases of the investigated structures in all patients, yielding a sensitivity of 73% and a specificity of 100%.

Conclusions

POCUS can be a reliable, valuable addition to physical examination of people with HIV and advanced immune suppression. The data support further exploration of POCUS in people with HIV in resource affluent settings.

Keywords

point-of-care-ultrasound HIV AIDS opportunistic diseases diagnostic accuracy value based health-care

Background

Ultrasound is a diagnostic tool that is increasingly used by clinicians from diverse specialties and has recently been added to the curriculum of internal medicine specialty training in various countries. However, its role in clinical practice for the internal medicine and infectious diseases specialist is not yet clearly defined. In settings endemic for tuberculosis (TB), the Fast Assessment with Sonography for HIV/TB (FASH) has been developed to assess people with HIV for signs of extrapulmonary TB.¹ The protocol includes an assessment of pericardial and pleural effusions, ascites, abdominal lymphadenopathy and splenic micro abscesses. It is widely implemented in Sub-Sahara Africa, due to its high clinical relevance, availability at the bedside and ease to learn for non-expert sonographers.² Other ultrasound applications relevant in people with HIV include lung ultrasound, which has been used to detect Pneumocystis jirovecii pneumonia (PJP)^3,4 and TB.^5–7 In addition, ultrasound can be used to assess the liver, spleen and kidneys for size and focal lesions which may indicate infections, lymphoma, or Kaposi sarcoma.^8,9 Despite the numerous indications and shown benefit for patient care, most studies on ultrasound in HIV patients have been performed in Africa. One retrospective study in Italy described typical ultrasound findings in a majority of people with HIV with extrapulmonary TB and therefore suggests integration of POCUS in the evaluation of extrapulmonary TB in resource affluent settings.¹⁰ However, no other studies have previously explored the value of ultrasound in HIV in resource affluent settings. In the Netherlands, there is still a significant percentage of people with HIV who present late in the course of their infection with cellular immunodeficiency due to a CD4 T-cell count below 350/mm³ and/or clinical signs of AIDS. In 2021, this was the case for 52% of men who have sex with men, 69% of other men and 58% of women with newly diagnosed HIV (57% of total HIV diagnoses).¹¹ The late presentations were associated with heterosexual transmission of HIV, older age, and Asian or African descent.¹² These patients are at high risk of having opportunistic diseases when they present to healthcare facilities and a significant proportion are from TB endemic areas.

Ultrasound may therefore help the clinician of not only resource restricted but also resource affluent settings in easily and rapidly detecting opportunistic diseases at the bedside or in the consultation room at the outpatient HIV clinic. With this study we determined the role of bedside POCUS in detecting opportunistic diseases in immune compromised people with HIV in a resource affluent setting.

Methods

We performed a prospective observational cohort study in the adult infectious diseases ward and outpatient HIV clinic of Erasmus University Medical Center. In the outpatient clinic, people with confirmed HIV infection and a CD4 T-cell count below 350 cells/mm3 were recruited. In addition, all people with HIV requiring admission to the hospital were eligible for participation. Patients were included between August 2020 to December 2021. Inability to sign informed consent was the only exclusion criterion. A formal sample size calculation was not performed as there are no previous data on POCUS in patients with HIV in our setting. However, based on the number of patients with a new HIV diagnosis and a CD4 T-cell count <350/mm³ from 2014-2020 in our hospital, we estimated that we needed to include a minimum of 37 patients to observe at least ten patients with an AIDS defining illness that may be detected by POCUS. The target inclusion number for sufficient AIDS defining illnesses was probably lower since admitted patients were expected to predominantly occur because of AIDS. Our research was performed in accordance with the Declaration of Helsinki and was approved by medical ethics committee of Erasmus Medical Centre, Rotterdam, the Netherlands. Written informed consent was obtained from all participants. This study was registered at ClinicalTrials.gov, NCT04246983

All patients were scanned using a standard protocol (Supplementary File 1) by clinicians with registered POCUS training (a 2-day course by Dutch Emergency Ultrasound and a 6-day course with a radiologist supervised examination from Boerhaave Continued Medical Education (CME) in Leiden for MH and CR respectively) followed by over 3 years of regular experience with point-of-care ultrasound in clinical practice (MH and CR). A Philips Lumify Convex (abdominal) probe was used to obtain most views. A linear probe was used to scan the spleen for micro-abscesses. Clinical data of enrolled participants related to the presence or absence of opportunistic diseases were available to the sonographer and recorded at the time of the ultrasound examination and until one year later during follow up in routine care with their own treating physician. Opportunistic diseases were defined as diagnosed by the treating physician who was aware of the POCUS results. Primary outcome was the number of HIV-related and unrelated conditions detected by POCUS at inclusion, the POCUS acceptance rate of patients, and interobserver variation in interpretation of ultrasound images. Secondary outcome was the number of opportunistic diseases observed during follow up and the added diagnostic value of POCUS to detect these in an early stage.

Descriptive statistics were applied to analyze baseline characteristics and ultrasound findings. Interobserver agreement was evaluated by evaluating the proportion similar interpretation of ultrasound images by a second reader (MH and CR) without knowledge of prior interpretation, on selected stored key images of each assessed organ system (by FASH and focussed ultrasound of lungs, kidneys, liver). Cohen’s kappa was estimated to describe interobserver agreement. The diagnostic test characteristics of POCUS to detect confirmed opportunistic diseases at the time of POCUS was assessed.

Results

Baseline characteristics are provided in Table 1. Of 34 included patients, 14 were admitted and the majority were male. Comorbidities were present in 18 patients, predominantly hypertension and/or cardiovascular diseases (n = 4), (history of) drug abuse (n = 4), and diabetes mellitus (n = 2). Median age was 53 (IQR 40-57), most patients were severely immune compromised with a median CD4 T-cell count of 115 cells/mm3 (IQR 32-248). A total of 16 and 15 patients had previously had an opportunistic disease or an opportunistic disease was present at the time of POCUS, respectively (Table 2). The opportunistic disease rate in the study cohort was 1.2/patient, comparable to the rate observed in a contemporary national cohort of people with HIV at risk for opportunistic disease in the Netherlands (1.5/patient).¹⁰

Table 1.

Baseline characteristics of the study population.

	Total (n = 34)	Admitted (n = 14)	Outpatient (n = 20)
Males, n (%)	27 (79)	12 (85)	15 (75)
Females, n (%)	6 (18)	2 (14)	4 (20)
Transgender, n (%)	1 (3)	0 (0)	1 (5)
Age, median (IQR)	53 (40-57)	54 (47-62)	49 (37-55)
CD4 T-cell count (cells/mm3), median (IQR)	115 (32-248)	125 (65-320)	115 (18-243)
Viral load (cp/ml), median (IQR)	727500 (100,850-1,907,500)	799000 (180,950-1,557,500)	585500 (100,850-1,917,500)
Current ART, n (%)	17 (50)	9 (64)	8 (40)
History of OD, n (%)	16 (47)	10 (71)	6 (30)
Symptoms of OD, n (%)	23 (68)	13 (93)	10 (50)
PE indicating OD, n (%)	21 (62)	12 (85)	9 (45)

Abbreviations: IQR: Interquartile range, ART: Antiretroviral therapy, OD: opportunistic disease, PE: physical examination

Table 2.

Observed opportunistic diseases in enrolled people with HIV from the study cohort before inclusion, at inclusion and during 1 year follow up after inclusion.

	Previous OD	Confirmed OD at time of POCUS	Confirmed OD during follow up
PJP	1	5	0
Oropharyngeal candida	6	2	0
Herpes simplex	2	0	0
Kaposi sarcoma	1	1	0
Pulmonary TB	1	0	0
Bacterial pneumonia	3	5	3
CMV pneumonia	2	0	0
Herpes zoster	2	0	2
Lymphoma	1	2	1
Toxoplasmosis	1	1	0
MAC	1	2	2
Skin condition^a	2	1	1
PML	0	0	1
Aspergillosis	0	1	0
Cryptococcemia	0	1	0
Total number of patients ^b	16	15	11

^aIncluding psoriasis, eczema, severe bacterial skin infection.

^bSome patients had more than one opportunistic disease.

Abbreviations: OD: opportunistic disease, POCUS: Point-of-care ultrasound, PJP: Pneumocystis jirovecii pneumonia, TB: tuberculosis, CMV: cytomegalovirus, MAC: Mycobacterium avium complex, PML: progressive multifocal leukoencephalopathy.

POCUS findings are summarised in Tables 3 and 4. Observers had a similar interpretation of POCUS in 97.5% of ultrasound images, with an estimated Cohen’s kappa of 0.75. An abnormality was observed in 21 patients, but only in 11 this was suggestive of opportunistic disease. In all of these 11 patients an opportunistic disease compatible with ultrasound findings was diagnosed. Overall, 11/15 patients with and 19/19 patients without any confirmed opportunistic disease at the time of POCUS were correctly identified yielding a sensitivity of 73%, specificity of 100%, positive predictive value of 100% and negative predictive value of 83%. Overall accuracy was 88% with 44% prevalent disease. No cases with a confirmed opportunistic disease at time of POCUS at an anatomical location accessible for ultrasound were missed by POCUS. No abnormalities falsely adjudicated as opportunistic diseases by POCUS were found. Two cases are described in more detail to illustrate the value of POCUS (Supplementary File 2, videos and figure). Four patients had opportunistic diseases that were not detected by ultrasound, but these were cases where a normal POCUS was expected given the location of the opportunistic illness and included a herpes zoster, cerebral toxoplasmosis, cryptococcal meningitis and Mycobacterium avium complex arthritis. No opportunistic diseases were observed in asymptomatic patients. In four symptomatic patients where physical examination was unremarkable, POCUS enabled the detection of a pleural effusion and pneumonia in one patient. In the other three patients no opportunistic illness was diagnosed despite the presence of symptoms at the time of ultrasound. These symptoms (weight loss and fatigue) were attributed to the HIV infection itself.

Table 3.

Point-of-care ultrasound findings in people with HIV.

	Total (n = 34)	Admitted (n = 14)	Outpatient (n = 20)
Ultrasound abnormal^a	21 (62)	13 (93)	8 (40)
Ultrasound suggesting OD^b	11 (32)	8 (57)	3 (15)
Confirmed OD at time of POCUS ^c,d	15 (44)	12 (86)	3 (15)
US suggesting OD and confirmed compatible OD at time of POCUS	11 (32)	8 (57)	3 (15)

^aMost common abnormalities were lung consolidations (n = 12), splenomegaly (n = 7), focal liver abnormalities (n = 6, mostly haemangiomas), pleural effusions (n = 5), and renal cysts (n = 4).

^bAbnormalities not suggestive of OD included liver/renal cysts, hemangiomas and subtle subpleural abnormalities.

^cMost common ODs were: Pneumocystis jirovecii pneumonia (n = 5), bacterial pneumonia (n = 5), Mycobacterium avium complex (n = 2) and lymphoma (n = 2).

^dPatients with a normal ultrasound and an OD were diagnosed with mycobacterium avium complex arthritis (n = 1), herpes zoster (n = 1), cryptococcal meningitis (n = 1), and cerebral toxoplasmosis (n = 1).

AbbreviationsOD: opportunistic disease, POCUS: Point-of-care ultrasound.

Table 4.

Positive ultrasound findings in patients with confirmed opportunistic disease.

Patient	Sex	Age	Diagnosis	Ultrasound findings
HIV-US-3	Male	75	Right sided pneumonia with pulmonary abscess	Right sided pleural effusion and consolidation (region R5)
HIV-US-8	Male	40	Right sided pneumonia with pulmonary abscess	Large right sided isoechoic mass (region R5)
HIV-US-16	Female	52	Bilateral bacterial pneumonia (Haemophilus influenzae)	Bilateral consolidations and subpleural nodules (regions R2, R5, R6, L4, L5 and L6)
HIV-US-17	Male	55	Pneumocystis jirovecii pneumonia	Bilateral consolidations (subtle C-pattern in regions R5, L1, L2), subpleural nodule in R3
HIV-US-18	Male	50	Pneumocystis jirovecii and Aspergillus pneumonia	Subpleural noduli and small right sided consolidations (C-pattern) in regions R3, R5, R6
HIV-US-19	Male	61	Diffuse large B-cell lymphoma	Abdominal/periportal lymph nodes, irregular texture of spleen without evident focal lesions or micro-abscesses
HIV-US-23 ^a	Male	57	Primary effusion lymphoma	Large pericardial effusion, bilateral pleural effusion and splenomegaly (149mm)
HIV-US-24	Male	42	Pneumocystis jirovecii pneumonia	Subpleural nodules in L2 and L6, small subpleural consolidation in R3
HIV-US-25	Male	54	Diffuse large B-cell lymphoma	Bilateral pleural effusion with basal consolidation (Regions R6 and L6), trace of pericardial effusion, abdominal lymphadenopathy
HIV-US-27^b	Male	57	Pneumocystis jirovecii pneumonia, diarrhea due to microsporidium infection	Subpleural nodules in all fields except L2, small right sided pleural effusion, free fluid in rectovesical pouch
HIV-US-32	Female	54	Pulmonary Mycobacterium avium complex ^c	Multiple subpleural nodules in all fields and small left sided subpleural consolidation (region L6)
HIV-US-34	Male	28	Kaposi sarcoma	Large left sided pleural effusion. Trace of right sided pleural effusion, and trace of free fluid in hepato- and splenorenal recess. Splenomegaly (14mm)

^aThis case is discussed in more detail, including video footage, in the supplement (Case 2). The ultrasound findings described here were only evident during a follow up scan.

^bThis case is discussed in more detail, including video footage, in the supplement (Case 1).

^cThis case was first classified as PJP by the treatment team based on a positive PCR in sputum and clinical likelihood despite atypical findings on imaging and a low LDH.

Eleven patients developed an opportunistic disease during follow up after median 143 days (IQR 46-214), most commonly bacterial pneumonia (n = 3), Mycobacterium avium complex (MAC) infection (n = 2) and herpes zoster (n = 2) (Table 2). In one of these patients ultrasound findings at inclusion were retrospectively considered to be compatible with the final diagnosis of pulmonary MAC four months later, but were first classified as PJP by the treatment team based on a positive PCR in sputum and clinical likelihood despite atypical findings on imaging and a low LDH. In the other ten patients POCUS did not assist in predicting the development of opportunistic diseases during follow up.

Discussion

We describe the POCUS findings in a cohort of people with late stage HIV in the Netherlands. A significant proportion of these patients had an opportunistic disease and the majority of these were detectable by ultrasound. Moreover, no opportunistic diseases in locations accessible for POCUS were missed and we found no opportunistic diseases in asymptomatic patients. POCUS appeared most valuable in symptomatic patients and provided relevant clinical clues not always detected by physical examination that aided in diagnostic and therapeutic management. We did not observe a role for POCUS in detecting opportunistic diseases not yet clinically apparent.

Late HIV presentation remains an important problem in many countries, even in the context of achieved 90-90-90 goals.¹¹ Therefore, the differential diagnostic consideration and timely recognition of opportunistic diseases remain essential for the infectious diseases specialist caring for people with HIV. Our findings illustrate that POCUS can assist in the detection of many opportunistic diseases. As it can be performed directly at the bedside or in the outpatient clinic, it can expedite the diagnostic process and direct additional investigations. POCUS is therefore a cost effective strategy that can easily be integrated during physical examination to optimize patient management. The European Federation of Internal Medicine has defined a number of POCUS core competencies and clinical setting for internists.¹³ We argue these should be integrated into the training of infectious diseases specialists and other medical specialists who treat people with HIV with a specific focus on recognising POCUS features compatible with opportunistic disease.

Our study was limited by the small sample size, but provides valuable data as very few publications exist on POCUS in patients with HIV in resource affluent settings. Reproducing these results in other hospitals involving more clinician-sonographers both newly trained and more experienced is relevant for assessing generalizability and refining diagnostic accuracy estimates. A previous study in Italy described typical sonographic findings in patients with extrapulmonary TB and advised POCUS screening to be integrated in diagnostic and screening algorithms for extrapulmonary TB in non-endemic settings.¹⁰ To the best of our knowledge this is the only study evaluating POCUS for a wide range of opportunistic diseases in a resource affluent setting. Although larger studies are needed to define its exact place in the diagnostic work-up, and how POCUS expedites diagnosing opportunistic diseases, the currently available data support a role for POCUS in HIV care if the skills to execute POCUS are present.

In conclusion, POCUS can be a valuable addition to physical examination of a symptomatic patient with HIV and advanced immune suppression. POCUS training, focusing on recognition of opportunistic diseases, should be considered in the curriculum of medical specialists who care for people with HIV.

Supplemental Material

Supplemental Material - Exploring the role of point-of-care ultrasound for people with HIV in a resource affluent setting: A prospective observational study

Supplemental Material for Exploring the role of point-of-care ultrasound for people with HIV in a resource affluent setting: A prospective observational study by Mischa Huson and Casper Rokx

Supplemental Material

Footnotes

Acknowledgements

We thank all participants in the study and our colleague internist-infectious diseases specialists at Erasmus University Medical Center for their help. In particular we acknowledge prof. dr. B.J.A. Rijnders for his advice.

Author’s contributions

MH and CR designed the study, included patients, analysed the data and wrote the manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was financially supported by Aidsfonds (project number P-54902). The funding body had no role in the design of the study, data collection, analysis, interpretation or writing of the manuscript.

ORCID iD

Casper Rokx

Data Availability Statement

The sets used during the study are available from the corresponding author on reasonable request.

Supplemental Material

Supplemental material for this article is available online.

References

Heller

Wallrauch

Goblirsch

, et al. Focused assessment with sonography for HIV-associated tuberculosis (FASH): a short protocol and a pictorial review. Crit Ultrasound J 2012; 4(1): 21. DOI: 10.1186/2036-7902-4-21.

Heller

Mtemang'ombe

Huson

, et al. Ultrasound for patients in a high HIV/tuberculosis prevalence setting: a needs assessment and review of focused applications for Sub-Saharan Africa. Int J Infect Dis 2017; 56: 229–236. DOI: 10.1016/j.ijid.2016.11.001.

Giordani

Tamarozzi

Kaminstein

, et al. Point-of-care lung ultrasound for diagnosis of Pneumocystis jirovecii pneumonia: notes from the field. Crit Ultrasound J 2018; 10(1): 8. DOI: 10.1186/s13089-018-0089-0.

Limonta

Monge

Montuori

, et al. Lung ultrasound in the management of pneumocystis pneumonia: A case series. Int J STD AIDS 2018; 12: 956462418797872. DOI: 10.1177/0956462418797872.

Hunter

Bélard

Janssen

, et al. Miliary tuberculosis: sonographic pattern in chest ultrasound. Infection 2016; 44(2): 243–246. DOI: 10.1007/s15010-015-0865-8.

Agostinis

Copetti

Lapini

, et al. Chest ultrasound findings in pulmonary tuberculosis. Trop Doct 2017; 47(4): 320–328.

Montuori

Casella

Casazza

, et al. Lung ultrasonography in pulmonary tuberculosis: a pilot study on diagnostic accuracy in a high-risk population. Eur J Intern Med 2019; 66: 29–34. DOI: 10.1016/j.ejim.2019.06.002.

Heller

Huson

Giordano

M-T

, et al. Human immunodeficiency virus (HIV) infection and tuberculosis (TB). In: Dietrich

(ed). EFSUMB (European Federation of Societies for ultrasound in medicine and Biology) course Book. 2nd ed., 2019.

Huson

MAM

Kumwenda

Gumulira

, et al. Ultrasound findings in Kaposi sarcoma patients: overlapping sonographic features with disseminated tuberculosis. Ultrasound J 2023; 15(1): 27. DOI: 10.1186/s13089-023-00323-8.

10.

Giordani

Brunetti

Binazzi

, et al. Extrapulmonary mycobacterial infections in a cohort of HIV-positive patients: ultrasound experience from Vicenza. Italy. Infection 2013; 41(2): 409–414. DOI: 10.1007/s15010-012-0336-4.

11.

van Sighem

Wit

FWNM

Boyd

, et al. HIV Monitoring Report 2022. Human immunodeficiency Virus (HIV) infection in The Netherlands. Amsterdam: Stichting HIV Monitoring, 2022.

12.

Schouten

van Velde

Snijdewind

, et al. [Late diagnosis of HIV positive patients in Rotterdam, the Netherlands: risk factors and missed opportunities]. Ned Tijdschr Geneeskd 2013; 157(15): A5731.

13.

Torres-Macho

Aro

Bruckner

, et al. Point-of-care ultrasound in internal medicine: a position paper by the ultrasound working group of the European federation of internal medicine. Eur J Intern Med 2020; 73: 67–71. DOI: 10.1016/j.ejim.2019.11.016.

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.

0.00 MB

0.85 MB