Made by DATEXIS (Data Science and Text-based Information Systems) at Beuth University of Applied Sciences Berlin
Deep Learning Technology: Sebastian Arnold, Betty van Aken, Paul Grundmann, Felix A. Gers and Alexander Löser. Learning Contextualized Document Representations for Healthcare Answer Retrieval. The Web Conference 2020 (WWW'20)
Funded by The Federal Ministry for Economic Affairs and Energy; Grant: 01MD19013D, Smart-MD Project, Digital Technologies
Traditionally, varicose veins were investigated using imaging techniques only if there was a suspicion of deep venous insufficiency, if they were recurrent, or if they involved the saphenopopliteal junction. This practice is not now widely accepted. People with varicose veins should now be investigated using lower limbs venous ultrasonography. The results from a randomised controlled trial on patients with and without routine ultrasound have shown a significant difference in recurrence rate and reoperation rate at 2 and 7 years of follow-up.
Clinical test that may be used include:
- Trendelenburg test–to determine the site of venous reflux and the nature of the saphenofemoral junction
Genetic tests are available for the "ENG", "ACVRL1" and "MADH4" mutations. Testing is not always needed for diagnosis, because the symptoms are sufficient to distinguish the disease from other diagnoses. There are situations in which testing can be particularly useful. Firstly, children and young adults with a parent with definite HHT may have limited symptoms, yet be at risk from some of the complications mentioned above; if the mutation is known in the affected parent, absence of this mutation in the child would prevent the need for screening tests. Furthermore, genetic testing may confirm the diagnosis in those with limited symptoms who otherwise would have been labeled "possible HHT" (see below).
Genetic diagnosis in HHT is difficult, as mutations occur in numerous different locations in the linked genes, without particular mutations being highly frequent (as opposed to, for instance, the ΔF508 mutation in cystic fibrosis). Sequence analysis of the involved genes is therefore the most useful approach (sensitivity 75%), followed by additional testing to detect large deletions and duplications (additional 10%). Not all mutations in these genes have been linked with disease.
Mutations in the "MADH4" gene is usually associated with juvenile polyposis, and detection of such a mutation would indicate a need to screen the patient and affected relatives for polyps and tumors of the large intestine.
Dilated submucosal veins are the most prominent histologic feature of esophageal varices. The expansion of the submucosa leads to elevation of the mucosa above the surrounding tissue, which is apparent during endoscopy and is a key diagnostic feature. Evidence of recent variceal hemorrhage includes necrosis and ulceration of the mucosa. Evidence of past variceal hemorrhage includes inflammation and venous thrombosis.
Identification of AVMs requires detailed medical imaging of the organs most commonly affected by these lesions. Not all AVMs cause symptoms or are at risk of doing so, and hence there is a degree of variation between specialists as to whether such investigations would be performed, and by which modality; often, decisions on this issue are reached together with the patient.
Lung AVMs may be suspected because of the abnormal appearance of the lungs on a chest X-ray, or hypoxia (low oxygen levels) on pulse oximetry or arterial blood gas determination. Bubble contrast echocardiography (bubble echo) may be used as a screening tool to identify abnormal connections between the lung arteries and veins. This involves the injection of agitated saline into a vein, followed by ultrasound-based imaging of the heart. Normally, the lungs remove small air bubbles from the circulation, and they are therefore only seen in the right atrium and the right ventricle. If an AVM is present, bubbles appear in the left atrium and left ventricle, usually 3–10 cardiac cycles after the right side; this is slower than in heart defects, in which there are direct connections between the right and left side of the heart. A larger number of bubbles is more likely to indicate the presence of an AVM. Bubble echo is not a perfect screening tool as it can miss smaller AVMs and does not identify the site of AVMs. Often contrast-enhanced computed tomography (CT angiography) is used to identify lung lesions; this modality has a sensitivity of over 90%. It may be possible to omit contrast administration on modern CT scanners. Echocardiography is also used if there is a suspicion of pulmonary hypertension or high-output cardiac failure due to large liver lesions, sometimes followed by cardiac catheterization to measure the pressures inside the various chambers of the heart.
Liver AVMs may be suspected because of abnormal liver function tests in the blood, because the symptoms of heart failure develop, or because of jaundice or other symptoms of liver dysfunction. The most reliable initial screening test is Doppler ultrasonography of the liver; this has a very high sensitivity for identifying vascular lesions in the liver. If necessary, contrast-enhanced CT may be used to further characterize AVMs. It is extremely common to find incidental nodules on liver scans, most commonly due to focal nodular hyperplasia (FNH), as these are a hundredfold times more common in HHT compared to the general population. FNH is regarded as harmless. Generally, tumor markers and additional imaging modalities are used to differentiate between FNH and malignant tumors of the liver. Liver biopsy is discouraged in people with HHT as the risk of hemorrhage from liver AVMs may be significant. Liver scans may be useful if someone is suspected of HHT, but does not meet the criteria (see below) unless liver lesions can be demonstrated.
Brain AVMs may be detected on computed tomography angiography (CTA or CT angio) or magnetic resonance angiography (MRA); CTA is better in showing the vessels themselves, and MRA provides more detail about the relationship between an AVM and surrounding brain tissue. In general, MRI is recommended. Various types of vascular malformations may be encountered: AVMs, micro-AVMs, telangiectasias and arteriovenous fistulas. If surgery, embolization, or other treatment is contemplated (see below), cerebral angiography may be required to get sufficient detail of the vessels. This procedure carries a small risk of stroke (0.5%) and is therefore limited to specific circumstances. Recent professional guidelines recommend that all children with suspected or definite HHT undergo a brain MRI early in life to identify AVMs that can cause major complications. Others suggest that screening for cerebral AVMs is probably unnecessary in those who are not experiencing any neurological symptoms, because most lesions discovered on screening scans would not require treatment, creating undesirable conundrums.
In emergency situations, care is directed at stopping blood loss, maintaining plasma volume, correcting disorders in coagulation induced by cirrhosis, and appropriate use of antibiotics such as quinolones or ceftriaxone. Blood volume resuscitation should be done promptly and with caution. The goal should be hemodynamic stability and hemoglobin of over 8 g/dl. Resuscitation of all lost blood leads to increase in portal pressure leading to more bleeding. Volume resuscitation can also worsen ascites and increase portal pressure. (AASLD guidelines)
Therapeutic endoscopy is considered the mainstay of urgent treatment. The two main therapeutic approaches are variceal ligation or banding and sclerotherapy.
In cases of refractory bleeding, balloon tamponade with a Sengstaken-Blakemore tube may be necessary, usually as a bridge to further endoscopy or treatment of the underlying cause of bleeding (usually portal hypertension). Esophageal devascularization operations such as the Sugiura procedure can also be used to stop complicated variceal bleeding. Methods of treating the portal hypertension include: transjugular intrahepatic portosystemic shunt, or a distal splenorenal shunt procedure or a liver transplantation.
Nutritional supplementation is not necessary if the patient is not eating for four days or less.
Terlipressin and octreotide for 1 to 5 days have also been used.
Since it is a rare disease, it remains a diagnosis of exclusion of other conditions with similar symptoms. The diagnosis is supported by the results of imaging studies such as computed tomography or magnetic resonance imaging, ultrasound of the abdomen (with or without doppler imaging) or intravenous urography.
Specialist vascular ultrasonographers should routinely look for left ovarian vein reflux in patients with lower limb varices especially if not associated with long or short saphenous reflux. The clinical pattern of varices differs between the two types of lower limb varices.
CT scanning is used to exclude abdominal or pelvic pathology. CT-Angiography/Venography can often demonstrate left ovarian vein reflux and image an enlarged left ovarian vein but is less sensitive and much more expensive than duplex Doppler ultrasound examination. Ultrasound requires that the ultrasonographer be experienced in venous vascular ultrasound and so is not always readily available. A second specialist ultrasound exam remains preferable to a CT scan.
As a wide range of pelvic and abdominal pathology can cause symptoms consistent with those symptoms due to left ovarian vein reflux, prior to embolisation of the left ovarian vein, a careful search for such diagnoses is essential. Consultation with general surgeons, gynaecologists, and possibly CT scanning should always be considered.
When Budd–Chiari syndrome is suspected, measurements are made of liver enzyme levels and other organ markers (creatinine, urea, electrolytes, LDH).
Budd–Chiari syndrome is most commonly diagnosed using ultrasound studies of the abdomen and retrograde angiography. Ultrasound may show obliteration of hepatic veins, thrombosis or stenosis, spiderweb vessels, large collateral vessels, or a hyperechoic cord replacing a normal vein. Computed tomography (CT) or magnetic resonance imaging (MRI) is sometimes employed although these methods are generally not as sensitive. Liver biopsy is nonspecific but sometimes necessary to differentiate between Budd–Chiari syndrome and other causes of hepatomegaly and ascites, such as galactosemia or Reye's syndrome.
Unlike oesophageal varices, rectal varices are less prone to bleeding, are less serious when a bleed does occur, and are easier to treat because of the more accessible location.
Typically, treatment consists of addressing the underlying portal hypertension. Some treatments include portosystemic shunting, ligation, and under-running suturing. Insertion of a transjugular intrahepatic portosystemic shunt (TIPS) has been shown to alleviate varices caused by portal hypertension. Successful treatment of portal hypertension that subsequently reduces anorectal varices provides a confirmation of the initial diagnosis, allowing for a distinction between varices and hemorrhoids, which would not have been alleviated by reduction of portal hypertension.
Diagnosis of gastric varices is often made at the time of upper endoscopy.
The Sarin classification of gastric varices identifies four different anatomical types of gastric varices, which differ in terms of treatment modalities.
Splenic infarction can be induced for the treatment of such conditions as portal hypertension or splenic injury. It can also be used prior to splenectomy for the prevention of blood loss.
The diagnosis of portopulmonary hypertension is based on hemodynamic criteria:
1. . Portal hypertension and/or liver disease (clinical diagnosis—ascites/varices/splenomegaly)
2. . Mean pulmonary artery pressure—MPAP > 25 mmHg at rest
3. . Pulmonary vascular resistance—PVR > 240 dynes s cm−5
4. . Pulmonary artery occlusion pressure— PAOP 12 mmHg where TPG = MPAP − PAOP.
The diagnosis is usually first suggested by a transthoracic echocardiogram, part of the standard pre-transplantation work-up. Echocardiogram estimated pulmonary artery systolic pressures of 40 to 50 mm Hg are used as a screening cutoff for PPH diagnosis, with a sensitivity of 100% and a specificity as high as 96%. The negative predictive value of this method is 100% but the positive predictive value is 60%. Thereafter, these patients are referred for pulmonary artery catheterization.
The limitations of echocardiography are related to the derivative nature of non-invasive PAP estimation. The measurement of PAP by echocardiogram is made using a simplified Bernoulli equation. High cardiac index and pulmonary capillary wedge pressures, however, may lead to false positives by this standard. By one institution’s evaluation, the correlation between estimated systolic PAP and directly measured PAP was poor, 0.49. For these reasons, right heart catheterization is needed to confirm the diagnosis.
Following diagnosis, mean survival of patients with PPH is 15 months. The survival of those with cirrhosis is sharply curtailed by PPH but can be significantly extended by both medical therapy and liver transplantation, provided the patient remains eligible.
Eligibility for transplantation is generally related to mean pulmonary artery pressure (PAP). Given the fear that those PPH patients with high PAP will suffer right heart failure following the stress of post-transplant reperfusion or in the immediate perioperative period, patients are typically risk-stratified based on mean PAP. Indeed, the operation-related mortality rate is greater than 50% when pre-operative mean PAP values lie between 35 and 50 mm Hg; if mean PAP exceeds 40-45, transplantation is associated with a perioperative mortality of 70-80% (in those cases without preoperative medical therapy). Patients, then, are considered to have a high risk of perioperative death once their mean PAP exceeds 35 mm_Hg.
Survival is best inferred from published institutional experiences. At one institution, without treatment, 1-year survival was 46% and 5-year survival was 14%. With medical therapy, 1-year survival was 88% and 5-year survival was 55%. Survival at 5 years with medical therapy followed by liver transplantation was 67%. At another institution, of the 67 patients with PPH from 1652 total cirrhotics evaluated for transplant, half (34) were placed on the waiting list. Of these, 16 (48%) were transplanted at a time when 25% of all patients who underwent full evaluation received new livers, meaning the diagnosis of PPH made a patient twice as likely to be transplanted, once on the waiting list. Of those listed for transplant with PPH, 11 (33%) were eventually removed because of PPH, and 5 (15%) died on the waitlist. Of the 16 transplanted patients with PPH, 11 (69%) survived for more than a year after transplant, at a time when overall one-year survival in that center was 86.4%. The three year post-transplant survival for patients with PPH was 62.5% when it was 81.02% overall at this institution.
Anorectal varices are the dilation of collateral submucosal vessels due to backflow in the veins of the rectum. Typically this occurs due to portal hypertension which shunts venous blood from the portal system through the portosystemic anastomosis present at this site into the systemic venous system. This can also occur in the oesophagus, causing oesophageal varices, and at the level of the umbilicus, causing caput medusae. Between 44% and 78% of patients with portal hypertension get anorectal varices.
Several studies have attempted to predict the survival of patients with Budd–Chiari syndrome. In general, nearly 2/3 of patients with Budd–Chiari are alive at 10 years. Important negative prognostic indicators include ascites, encephalopathy, elevated Child-Pugh scores, elevated prothrombin time, and altered serum levels of various substances (sodium, creatinine, albumin, and bilirubin). Survival is also highly dependent on the underlying cause of the Budd–Chiari syndrome. For example, a patient with an underlying myeloproliferative disorder may progress to acute leukemia, independently of Budd–Chiari syndrome.
A varix (pl. varices) is an abnormally dilated vessel with a tortuous course. Varices usually occur in the venous system, but may also occur in arterial or lymphatic vessels.
Examples of varices include:
- Varicose veins, large tortuous veins usually found on legs
- Sublingual varices
- Esophageal varices, commonly stemming from cirrhosis of the liver, also known as oesophageal varicose
- Gastric varices, commonly stemming from cirrhosis of the liver
- Intestinal varices
- Scrotal varices
- Vulvar varices
- Pelvic varices
- Vesical varices, varicose veins associated with the urinary bladder
- Rectal varices, which can be similar to external haemorrhoids
Spider angiomas are asymptomatic and usually resolve spontaneously. This is common in the case of children, although they may take several years to disappear. If the spider angiomas are associated with pregnancy, they may resolve after childbirth. In women taking oral contraceptives, they may resolve after stopping these contraceptives. The spider angiomas associated with liver disease may resolve when liver function increases or when a liver transplant is performed.
For spider angiomas on the face, techniques such as electrodesiccation and laser treatment can be used to remove the lesion. There is a small risk of a scar, although the results are generally good. Spider angiomas can recur after treatment.
The Cognard et al. Classification correlates venous drainage patterns with increasingly aggressive neurological clinical course.
Treatment consists of painkillers and surgical ablation of the dilated vein. This can be accomplished with open abdominal surgery (laparotomy) or keyhole surgery (laparoscopy). Recently, the first robot-assisted surgery was described.
Another approach to treatment involves catheter-based embolisation, often preceded by phlebography to visualise the vein on X-ray fluoroscopy.
Ovarian vein coil embolisation is an effective and safe treatment for pelvic congestion syndrome and lower limb varices of pelvic origin. Many patients with lower limb varices of pelvic origin respond to local treatment i.e. ultrasound guided sclerotherapy. In those cases, ovarian vein coil embolisation should be considered second line treatment to be used if veins recur in a short time period i.e. 1–3 years. This approach allows further pregnancies to proceed if desired. Coil embolisation is not appropriate if a future pregnancy is possible. This treatment has largely superseded operative options.
Coil embolisation requires exclusion of other pelvic pathology, expertise in endovascular surgery, correct placement of appropriate sized coils in the pelvis and also in the upper left ovarian vein, careful pre- and post-procedure specialist vascular ultrasound imaging, a full discussion of the procedure with the patient i.e. informed consent. Complications, such as coil migration, are rare but reported. Their sequelae are usually minor.
If a Nutcracker compression (see below) is discovered, stenting of the renal vein should be considered before embolization of the ovarian vein. Reducing outflow obstruction should always be the main objective.
Manual carotid self compression is a controversial treatment for DAVF. Patients using this method are told to compress the carotid with the opposite hand for approximately 10 minutes daily, and gradually increasing the frequency and duration of compression. Currently, it is unclear whether this method is an effective therapy.
Initial treatment of bleeding from gastric varices focuses on resuscitation, much as with esophageal varices. This includes administration of fluids, blood products, and antibiotics.
The results from the only two randomized trials comparing band ligation vs cyanoacrylate suggests that endoscopic injection of cyanoacrylate, known as gastric variceal obliteration or GVO is superior to band ligation in preventing rebleeding rates. Cyanoacrylate, a common component in 'super glue' is often mixed 1:1 with lipiodol to prevent polymerization in the endoscopy delivery optics, and to show on radiographic imaging. GVO is usually performed in specialized therapeutic endoscopy centers. Complications include sepsis, embolization of glue, and obstruction from polymerization in the lumen of the stomach.
Other techniques for refractory bleeding include:
- Transjugular intrahepatic portosystemic shunts (TIPS)
- Balloon occluded retrograde transvenous obliteration techniques (BORTO)
- Gastric variceal ligation, although this modality is falling out of favour
- Intra-gastric balloon tamponade as a bridge to further therapy
- a caveat is that a larger balloon is required to occupy the fundus of the stomach where gastric varices commonly occur
- Liver transplantation
Portal hypertensive gastropathy can also be treated with endoscopic treatment delivered through a fibre-optic camera into the stomach. Argon plasma coagulation and electrocautery have both been used to stop bleeding from ectatic vessels, and to attempt to obliterate the vessels, but have limited utility if the disease is diffuse.
Transjugular intrahepatic portosystemic shunt procedures, or TIPS involve decompressing the portal vein by shunting a portal venule to a lower pressure systemic venule, under guidance with fluoroscopy. Since it treats the root cause of portal hypertension gastropathy, it has been putatively used for the condition. The literature reports suggest both regression of portal hypertensive gastropathy on endoscopic images and improvement in bleeding after TIPS.
Finally, cryotherapy involves the use of pressurized carbon dioxide administered through the endoscope to freeze and destroy tissue in a focal area. It is being studied for the treatment of portal hypertensive gastropathy.
Congestion of the mucosa in other parts of the gastrointestinal tract can also be seen in portal hypertension. When the condition involves the colon, it is termed "portal hypertensive colopathy".
Hemorrhoids are typically diagnosed by physical examination. A visual examination of the anus and surrounding area may diagnose external or prolapsed hemorrhoids. A rectal exam may be performed to detect possible rectal tumors, polyps, an enlarged prostate, or abscesses. This examination may not be possible without appropriate sedation because of pain, although most internal hemorrhoids are not associated with pain. Visual confirmation of internal hemorrhoids may require anoscopy, insertion of a hollow tube device with a light attached at one end. The two types of hemorrhoids are external and internal. These are differentiated by their position with respect to the dentate line. Some persons may concurrently have symptomatic versions of both. If pain is present, the condition is more likely to be an anal fissure or an external hemorrhoid rather than an internal hemorrhoid.
The diagnosis of hemosuccus pancreaticus can be difficult to make. Most patients who develop bleeding in the gastrointestinal tract have endoscopic procedures done to visualize the bowel in order to find and treat the source of the bleeding. With hemosuccus, the bleeding is coming from the pancreatic duct which enters into the first part of the small intestine, termed the duodenum. Typical gastroscopes used to visualize the esophagus, stomach and duodenum are designed with fiber-optic illumination that is directed in the same direction as the endoscope, meaning that visualization is in the forward direction. However, the pancreatic duct orifice is located on the side of the duodenum, meaning that it can be missed on forward-viewing endoscopy. A side-viewing endoscope (known as a "duodenoscope", or "side-viewer") used for endoscopic retrograde cholangiopancreatography (ERCP), a procedure to visualize the bile ducts and pancreatic duct on fluoroscopy, can be used to localize the bleeding to the pancreatic duct. It can be confused with bleeding from the common bile duct on endoscopy, leading to the term "pseudohematobilia".
Liver function test is normal apart from an increased serum bilirubin in the event of pancreaticobiliary reflux. Serum amylase is normal outside episodes of acute pancreatitis. It is difficult to diagnose HP because the bleeding is usually intermittent. Endoscopy is essential in ruling out other causes of upper gastrointestinal bleeding and in rare cases; active bleeding can be seen from the duodenal ampulla. Even though endoscopy may be normal, it helps to rule out other causes of upper digestive bleeding (erosive gastritis, peptic ulcers, and oesophageal and gastric fundus varices, etc.). Ultrasonography can be used to visualize pancreatic pseudocysts or aneurysm of the peripancreatic arteries. Doppler ultrasound or dynamic ultrasound has been reported to be diagnostic. Contrast-enhanced CT is an excellent modality for demonstrating the pancreatic pathology and can also demonstrate features of chronic pancreatitis, pseudocysts, and pseudoaneurysms. On precontrast CT, the characteristic finding of clotted blood in the pancreatic duct, known as the sentinel clot, is seldom seen. Computed tomography may show simultaneous opacification of an aneurysmal artery and pseudocyst or persistence of contrast within a pseudocyst after the arterial phase. Again, these findings are only suggestive of the diagnosis. Ultimately, angiography is the diagnostic reference standard. Angiography identifies the causative artery and allows for delineation of the arterial anatomy and therapeutic intervention.