The spleen is the largest lymphatic organ, harboring many blood and immune cells, including around one-third of the total available platelets. It functions to 1) filter out particulates from the blood such as microorganisms, cell fragments, and old or deformed blood cells, 2) host macrophages that can recognize and remove circulating cells that are coated with immunoglobulin or complement proteins, 3) hold a reserve volume of blood cells (splenic volume decreases during exercise), 4) host and mature B lymphocytes, T lymphocytes, and plasma cells, which generate the body’s humoral immune response, and 5) sometimes host hematopoiesis (blood cell production) when the bone marrow is insufficient. Without a functioning spleen, patients become more vulnerable to infections that require a humoral immune response and may develop thrombocytosis, leukocytosis, and reduced hemolysis.
Image 11.1 Normal left upper quadrant longitudinal view of the spleen.
A normal spleen’s size can vary based on height and sex and fluctuate ±10-20% between measurements. Identifying an enlarged spleen may be insignificant but could help confirm a suspected disease process or trigger investigations that identify a life-threatening condition. The following may cause splenomegaly.
The spleen resides typically under the ribs and is soft and easily pushed away when exposed. So, even when patients are thin, an enlarged spleen can still be difficult to palpate, and physical exam maneuvers to detect splenomegaly lack accuracy and precision.1 However, sonographic measurement is often considered the reference standard. Still, if PoCUS simply replaced the physical exam to detect splenomegaly, it may expose more patients than necessary to the potentially harmful effects of downstream testing and treatments. Therefore, it is prudent to limit PoCUS assessment for splenomegaly to those with an abnormal physical exam or when an associated condition is already suspected. To image the spleen, use a phased array or curvilinear transducer in the LUQ near the mid-axillary line and achieve a view that cuts the spleen at its longest oblique cranial-caudal axis. Freeze the optimal view and measure the longest axis. Traditionally, 12-13cm is the upper limit of normal, but consider determining the specific upper limit for the patient’s height and sex, which may be over 14.5cm in tall men.2 Unfortunately, the upper limit of normal in children is more complicated and an active area of research. Consider only using PoCUS in small children to confirm splenomegaly by determining if a palpable enlarged spleen is actually the spleen.3
Image 11.2 Abdominal left upper quadrant longitudinal views showing splenomegaly measuring >18cm (A) and an even larger spleen (B).
Normally, the spleen is homogenous, but PoCUS may also reveal focal splenic lesions. These may represent abscesses, cysts, post-traumatic pseudocysts, hematomas, hemangiomas, lymphangiomas, hamartomas, and primary vascular tumors such as hemangiosarcoma. A particularly tender spleen suggests infarction, rupture, or acute infection. Be cautious not to mistake adjacent gastric or colonic tissue as a splenic lesion. A splenic abscess is uncommon and usually associated with fever, bacteremia, LUQ tenderness, and sometimes free fluid in the peritoneum or adjacent pleural space. Diffuse hypoechoic lesions, termed microabscesses, can occur in the context of tuberculosis and other conditions discussed below. Scattered calcifications may represent prior granulomatous infections, phleboliths (vein stones), amyloidosis, sickle cell disease, and lymphoma (after treatment). Hyperechoic calcifications may also occur around cysts, infarcts, hemangiomas, lymphangiomas, or splenic artery aneurysms. It is rare for tumors to metastasize to the spleen without other significant spread. If a focal splenic lesion is suspected, spend additional effort ensuring that it is indeed within the spleen. This can be difficult, especially when the splenic view is suboptimal. Consider reviewing images with a colleague or consultant and obtain consultative imaging.
The basic workup for splenomegaly assesses for signs and symptoms of specific causes discussed above, focusing on identifying lymphadenopathy, signs of portal hypertension, and abnormalities in the complete blood count (CBC). Cytopenias suggest venous congestion with hypersplenism, blood infection, autoimmune disease, or congenital blood disorders. Consider performing a peripheral smear, blood cultures, tests for autoantibodies or specific infections, and imaging to assess the liver and associated vasculature and identify accessible tissue to biopsy, such as enlarged lymph nodes. Consider bone marrow biopsy or referral to hematology whenever the CBC or peripheral smear shows abnormal lymphocytes or unexplained leukocytosis, bandemia, polycythemia, or thrombocytosis.
It may be appropriate to simply observe patients with mild splenomegaly who lack significant history, exam, and CBC findings. If electing to observe, re-evaluate after a year and educate the patient to monitor for new symptoms, especially weight loss and constitutional symptoms. Lastly, given an increased risk of potentially fatal splenic rupture, it is prudent to advise patients with splenomegaly to avoid falls, heavy lifting, or high-risk contact sports for a few months or longer if the associated condition and splenomegaly persist.4
Mycobacterium tuberculosis (TB) thrives when the immune system is weak. Risk factors for active disease include malnutrition, diabetes mellitus, steroid/immunosuppressant use, low vitamin D, end-stage kidney disease, chronic lung disease, and, most notably, HIV infection. At any CD4 count, but especially when reduced, HIV is associated with the developing active TB infection. TB also appears to promote progression of HIV and reduce the efficacy of antiretroviral therapy.5,6 This may be why active TB in HIV is often 1) disseminated via hematogenous spread to multiple organs, 2) associated with a low CD4 count, and 3) lacks typical findings specific for tuberculosis infection, like cavitary lesions or upper lobe infiltrates on chest radiography and positive sputum microscopy.
Where endemic, TB may be responsible for close to 40% of HIV-associated deaths in adults. Around 90% of these TB and HIV-related deaths involve disseminated or ‘miliary’ TB, and close to half may be undiagnosed at the time of death.7 In patients with HIV in TB endemic regions, chest radiography, fluid adenosine deaminase concentration, urine Lipoarabinomannan (LAM), and sputum microscopy, culture, or PCR testing can all confirm TB but are often unavailable and on their own are unable to rule out disseminated TB (dTB). Thus, a high index of suspicion for dTB is essential in regions with a high burden of HIV-associated TB, and local standards of care may include empiric treatment protocols where clinical suspicion bolstered by PoCUS findings may be sufficient to initiate prompt 1) empiric treatment when seriously ill,8 or 2) invasive testing that may not have otherwise occurred.9 Also, a lack of PoCUS findings in conjunction with a lack of other findings such as fever or abnormal chest radiograph may help clinicians avoid giving unnecessary TB treatment (TBT).10 PoCUS users should follow local standards of care and incorporate all data when determining a diagnosis, as PoCUS alone is insufficient to rule in or rule out TB.11
Non-PoCUS findings that increase the likelihood of TB with HIV include 1) cough of any type or duration, 2) weight loss or low BMI (lower is more specific), 3) anemia not otherwise explained (lower is more specific), and 4) fever not otherwise explained (higher is more specific). Also, note if there has been no improvement with empiric treatment for pneumonia or failure to improve nutritional status after starting antiretroviral therapy (ART) for HIV.
Chest radiography is an essential part of assessing for TB. PoCUS findings of focal or diffuse B-lines and consolidations, especially if isolated to an upper lobe, may be considered surrogates for abnormalities on chest radiography. However, thoracic ultrasound has not been as extensively studied in the context of TB diagnosis and may be more sensitive than chest radiographs. Thus, subtle PoCUS findings may be less specific. Attempt to differentiate chest imaging abnormalities from those due to heart failure, Pneumocystis pneumonia (often presents with significant shortness of breath and hypoxia), other bacterial pneumonia, Kaposi sarcoma (may have skin/oral lesion, lung mass, and not respond to TBT), Cryptococcus infection (serum antigen testing may be positive), chronic lung disease, and other inflammatory lung disease (does not improve with TBT).
Image 11.3 (A) Chest radiograph revealing diffuse bilateral micronodular opacities from the hematogenous spread of tuberculosis in a patient with HIV. (B) Ultrasound scanning of the same patient revealed diffuse B-lines. (C) Pulmonary consolidation in the right upper lobe of another patient with HIV and confirmed disseminated TB.
Pleural and pericardial effusions due to tuberculosis are often exudative but less so in advanced HIV. Thick, purulent fluid is not typically due to tuberculosis in this setting. PoCUS can help identify TB pericarditis and guide pericardiocentesis, which is especially important when pericardial tamponade is suspected, as discussed in the chapter on pericardial disease. Additional glucocorticoid therapy may also improve outcomes in patients with TB pericarditis, who risk developing subsequent tamponade and constrictive pericarditis.12 Finding ascites is more specific if it has septations/inflammatory stranding or occurs with abdominal lymphadenopathy. Standard fluid analysis and microbiologic testing are often insensitive and nonspecific for TB, but adenosine deaminase concentration (if available) can help determine if TB is the etiology.13
Image 11.4 (A) Subcostal cardiac view showing a large pericardial effusion in a patient with persistent cough and weight loss in an area where TB is endemic. (B) Left mid-axillary view of a large pleural effusion in a patient with cough, weight loss, and new HIV diagnosis. (C) Right upper quadrant view revealing abdominal free fluid in a patient with HIV and cough that had not improved on antibiotics with positive urine LAM testing. (D) A transverse pelvic view revealing complex pelvic free fluid in a patient with fever and weight loss who subsequently improved of TB treatment. (E) A parasternal short-axis view revealing a large left pleural and pericardial effusion in a patient with HIV and fever, weight loss, respiratory distress, and hypotension that immediately improved following thoracentesis.
The lymph nodes are a common extrapulmonary site for TB infection. Involved lymph nodes are usually hypoechoic and enlarged. In HIV-associated dTB, the abdominal lymph nodes are often disproportionately enlarged. The para-aortic lymph nodes are normally invisible or just a few small round hypodensities, all <1 cm in diameter. Para-aortic lymphadenopathy is present when there are multiple ball-like hypodensities >1cm adjacent to the aorta and IVC just below the epigastrium or subxiphoid area. Sometimes, enlarged lymph nodes converge into a hypoechoic mass. Enlarged nodes can even exert a mass effect on local vasculature.14,15 Abdominal lymphadenopathy (LAD) with para-aortic lymph nodes >1 cm diameter is a specific finding for dTB in at-risk patients in endemic settings.10,16 Other causes of abdominal lymphadenopathy can include lymphoma (more often involves inguinal and iliac nodes and less often with ascites),17 Kaposi sarcoma (may have skin/oral lesion, lung mass, and not respond to TBT), other malignancy, Cryptococcus infection (serum antigen testing may be positive), other infection, and sarcoidosis. PoCUS can guide diagnostic aspiration of enlarged lymph nodes.18 This can help identify lymphoma and other cancers, though partial surgical biopsies can lead to non-healing wounds.
Image 11.5 (A) Normal longitudinal abdominal aortic view. (B) Para-aortic lymphadenopathy, where multiple hypoechoic lymph nodes are visible (✶) in a patient with HIV and positive urine LAM testing. (C, D) Another example of para-aortic lymphadenopathy but in a patient without HIV or another risk factor for disseminated TB, necessitating stronger consideration for an alternative cause such as lymphoma.
Disseminated TB may enlarge the spleen or liver, but more specifically, it can form scattered hypoechoic lesions called microabscesses. This finding lacks sensitivity but is specific to TB in at-risk patients in endemic settings.10,16,19 To improve sensitivity, look at the spleen carefully with optimal gain, minimal depth, and reduced room lighting, and use a higher-frequency linear transducer. The characteristic lesions are hypoechoic, usually 5-10mm in diameter, and scattered evenly throughout the spleen. Other causes of hypoechoic splenic lesions include Cryptococcus, nocardiosis, cat-scratch disease, leishmaniasis, sarcoidosis, lymphoma, metastatic cancer, and other infections.
Image 11.6 A spleen with defuse hypoechoic lesions termed microabscesses in a patient with HIV and disseminated tuberculosis. Ideally, examine the spleen with a linear transducer to avoid overlooking this subtle finding.
How these findings affect diagnosis and treatment decisions depends on the pretest probability and local standards of care. Griesel et al. found that PoCUS findings of 1) abdominal LAD >1cm, 2) splenic microabscesses, and 3) ascites, pleural effusion, or pericardial effusion appeared to be independent predictors of TB in patients with HIV who were hospitalized with cough and either fever, tachycardia, tachypnea, or severe weakness. If all three of these PoCUS findings were present, the +LR was 17.9. If there were two of three, then the +LR was 4.3. If there was any one of three, then the +LR was 2.4. By comparison, in the same group of patients, a chest radiograph report saying tuberculosis “likely” had a +LR of 3.4.20 Conversely, Ndege et al. reported a prospective study of PoCUS findings in consecutive adult patients presenting to a referral hospital or clinic with suspected TB. Over half were HIV+, the majority with suspected dTB. TB was confirmed in over half of the patients, and less than 30% were considered negative after extensive testing and six months of follow-up. They showed that finding no fever, a normal chest x-ray, and no PoCUS findings concerning for TB had a negative predictive value of 95%.10 Thus, in this cohort, the probability of TB dropped from >50% to near 5%. While further testing would still be appropriate, this reduced likelihood could help justify withholding empiric TB treatment.
Acutely ill patients with dTB require TB treatment as soon as possible. PoCUS may identify specific findings of dTB and allow for earlier empiric treatment in patients with HIV and suspected TB.8 Be cautious about waiting 3-5 days to determine if empiric antibiotics for bacterial pneumonia will be sufficient to treat the patient’s illness, and consider avoiding quinolone antibiotics like levofloxacin whenever TB is possible, as they can partially treat the TB infection and may be required to treat drug-resistant TB. It may be appropriate to wait for confirmatory TB testing in patients without any danger signs. Still, avoid significantly delaying treatment in patients with suspected HIV-associated dTB. Patients with HIV also require effective ART, and active TB may be a risk factor for inadequate ART. When prescribing TB treatment, follow proper weight-based dosing, appropriately modify the regimen when the creatinine clearance is low, and assess for interacting medications, as TB drugs can affect the absorption/metabolism of many other drugs and vice versa. Also, educate patients on the need to avoid alcohol consumption, to ensure daily compliance despite the possibility of worsening before improving, and to attend follow-up care, which ensures that the drugs are tolerated and effective.
In addition, isolate patients with suspected TB while ensuring they get necessary supportive care, including oxygen, fluids/electrolytes, nutrition, and physical rehabilitation. Steroids may be used for meningitis and pericarditis. Use PoCUS to guide drainage of symptomatic fluid collections. Patients should be enrolled into a protocolized monitoring system to regularly assess for side effects and the effectiveness of both TB and HIV treatments. Consider repeating PoCUS after three months of treatment, as TB-related findings usually resolve, and their continued presence may indicate non-adherence, drug resistance, chronic disease, IRIS, or an alternative diagnosis.21,22