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
A physician's overall impression is most effective in initially making the diagnosis. Single factors are much less useful.
Methods used in laboratory diagnosis include culturing of nasopharyngeal swabs on a nutrient medium (Bordet-Gengou medium), polymerase chain reaction (PCR), direct fluorescent antibody (DFA), and serological methods (e.g. complement fixation test). The bacteria can be recovered from the person only during the first three weeks of illness, rendering culturing and DFA useless after this period, although PCR may have some limited usefulness for an additional three weeks.
Serology may be used for adults and adolescents who have already been infected for several weeks to determine whether antibody against pertussis toxin or another virulence factor of "B. pertussis" is present at high levels in the blood of the person. By this stage, they have been contagious for some weeks and may have spread the infection to many people. Because of this, adults, who are not in great danger from pertussis, are increasingly being encouraged to be vaccinated.
A similar, milder disease is caused by "B. parapertussis".
The primary method of prevention for pertussis is vaccination. Evidence is insufficient to determine the effectiveness of antibiotics in those who have been exposed, but are without symptoms. Preventive antibiotics, however, are still frequently used in those who have been exposed and are at high risk of severe disease (such as infants).
Many cases of croup have been prevented by immunization for influenza and diphtheria. At one time, croup referred to a diphtherial disease, but with vaccination, diphtheria is now rare in the developed world.
An oral whole cell nontypeable Haemophilus influenzae vaccine may protect against the disease, but "the evidence is mixed".
The most commonly used system for classifying the severity of croup is the Westley score. It is primarily used for research purposes rather than in clinical practice. It is the sum of points assigned for five factors: level of consciousness, cyanosis, stridor, air entry, and retractions. The points given for each factor is listed in the adjacent table, and the final score ranges from 0 to 17.
- A total score of ≤ 2 indicates "mild" croup. The characteristic barking cough and hoarseness may be present, but there is no stridor at rest.
- A total score of 3–5 is classified as "moderate" croup. It presents with easily heard stridor, but with few other signs.
- A total score of 6–11 is "severe" croup. It also presents with obvious stridor, but also features marked chest wall indrawing.
- A total score of ≥ 12 indicates impending respiratory failure. The barking cough and stridor may no longer be prominent at this stage.
85% of children presenting to the emergency department have mild disease; severe croup is rare (<1%).
Vaccination helps prevent bronchopneumonia, mostly against influenza viruses, adenoviruses, measles, rubella, streptococcus pneumoniae, haemophilus influenzae, diphtheria, bacillus anthracis, chickenpox, and bordetella pertussis.
Antibiotics do not help the many lower respiratory infections which are caused by parasites or viruses. While acute bronchitis often does not require antibiotic therapy, antibiotics can be given to patients with acute exacerbations of chronic bronchitis. The indications for treatment are increased dyspnoea, and an increase in the volume or purulence of the sputum. The treatment of bacterial pneumonia is selected by considering the age of the patient, the severity of the illness and the presence of underlying disease. Amoxicillin and doxycycline are suitable for many of the lower respiratory tract infections seen in general practice.
A physical examination will often reveal decreased intensity of breath sounds, wheezing, rhonchi, and prolonged expiration. Most physicians rely on the presence of a persistent dry or wet cough as evidence of bronchitis.
A variety of tests may be performed in patients presenting with cough and shortness of breath:
- A chest X-ray is useful to exclude pneumonia which is more common in those with a fever, fast heart rate, fast respiratory rate, or who are old.
- A sputum sample showing neutrophil granulocytes (inflammatory white blood cells) and culture showing that has pathogenic microorganisms such as "Streptococcus" species.
- A blood test would indicate inflammation (as indicated by a raised white blood cell count and elevated C-reactive protein).
Antibiotics are given to treat any bacterial infection present. Cough suppressants are used if the cough is not productive. NSAIDs are often given to reduce fever and upper respiratory inflammation. Prevention is by vaccinating for canine adenovirus, distemper, parainfluenza, and "Bordetella". In kennels, the best prevention is to keep all the cages disinfected. In some cases, such as "doggie daycares" or nontraditional playcare-type boarding environments, it is usually not a cleaning or disinfecting issue, but rather an airborne issue, as the dogs are in contact with each other's saliva and breath. Although most kennels require proof of vaccination, the vaccination is not a fail-safe preventative. Just like human influenza, even after receiving the vaccination, a dog can still contract mutated strains or less severe cases.
Dogs will typically recover from kennel cough within a few weeks. However, secondary infections could lead to complications that could do more harm than the disease itself. Several opportunistic invaders have been recovered from the respiratory tracts of dogs with kennel cough, including Streptococcus, Pasteurella, Pseudomonas, and various coliforms. These bacteria have the potential to cause pneumonia or sepsis, which drastically increase the severity of the disease. These complications are evident in thoracic radiographic examinations. Findings will be mild in animals affected only by kennel cough, while those with complications may have evidence of segmental atelectasis and other severe side effects.
The diagnosis is typically made by clinical examination. Chest X-ray is sometimes useful to exclude bacterial pneumonia, but not indicated in routine cases.
Testing for the specific viral cause can be done but has little effect on management and thus is not routinely recommended. RSV testing by direct immunofluorescence testing on nasopharyngeal aspirate had a sensitivity of 61% and specificity of 89%. Identification of those who are RSV-positive can help for: disease surveillance, grouping ("cohorting") people together in hospital wards to prevent cross infection, predicting whether the disease course has peaked yet, reducing the need for other diagnostic procedures (by providing confidence that a cause has been identified).
Infants with bronchiolitis between the age of two and three months have a second infection by bacteria (usually a urinary tract infection) less than 6% of the time. Preliminary studies have suggested that elevated procalcitonin levels may assist clinicians in determining the presence of bacterial coinfection, which could prevent unnecessary antibiotic use and costs.
In patients managed in the community, determining the causative agent is not cost-effective and typically does not alter management. For people who do not respond to treatment, sputum culture should be considered, and culture for "Mycobacterium tuberculosis" should be carried out in persons with a chronic productive cough. Testing for other specific organisms may be recommended during outbreaks, for public health reasons. In those hospitalized for severe disease, both sputum and blood cultures are recommended, as well as testing the urine for antigens to "Legionella" and "Streptococcus". Viral infections can be confirmed via detection of either the virus or its antigens with culture or polymerase chain reaction (PCR), among other techniques. The causative agent is determined in only 15% of cases with routine microbiological tests.
A chest radiograph is frequently used in diagnosis. In people with mild disease, imaging is needed only in those with potential complications, those not having improved with treatment, or those in which the cause is uncertain. If a person is sufficiently sick to require hospitalization, a chest radiograph is recommended. Findings do not always match the severity of disease and do not reliably separate between bacterial infection and viral infection.
X-ray presentations of pneumonia may be classified as lobar pneumonia, bronchopneumonia (also known as lobular pneumonia), and interstitial pneumonia. Bacterial, community-acquired pneumonia classically show lung consolidation of one lung segmental lobe, which is known as lobar pneumonia. However, findings may vary, and other patterns are common in other types of pneumonia. Aspiration pneumonia may present with bilateral opacities primarily in the bases of the lungs and on the right side. Radiographs of viral pneumonia may appear normal, appear hyper-inflated, have bilateral patchy areas, or present similar to bacterial pneumonia with lobar consolidation. Radiologic findings may not be present in the early stages of the disease, especially in the presence of dehydration, or may be difficult to be interpreted in the obese or those with a history of lung disease. A CT scan can give additional information in indeterminate cases. Lung ultrasound may also be useful in helping to make the diagnosis.
Chest radiographs (X-ray photographs) often show a pulmonary infection before physical signs of atypical pneumonia are observable at all.
This is occult pneumonia. In general, occult pneumonia is rather often present in patients with pneumonia and can also be caused by "Streptococcus pneumoniae", as the decrease of occult pneumonia after vaccination of children with a pneumococcal vaccine suggests.
Infiltration commonly begins in the perihilar region (where the bronchus begins) and spreads in a wedge- or fan-shaped fashion toward the periphery of the lung field. The process most often involves the lower lobe, but may affect any lobe or combination of lobes.
Diagnosis is typically based on a person's signs and symptoms. The color of the sputum does not indicate if the infection is viral or bacterial. Determining the underlying organism is typically not needed. Other causes of similar symptoms include asthma, pneumonia, bronchiolitis, bronchiectasis, and COPD. A chest X-ray may be useful to detect pneumonia.
Another common sign of bronchitis is a cough which lasts ten days to three weeks. If the cough lasts a month or a year it may be chronic bronchitis. In addition to having a cough a fever may be present. Acute bronchitis is normally caused by a viral infection. Typically these infections are rhinovirus, para influenza, or influenza. No specific testing is normally needed to diagnose acute bronchitis.
Mycoplasma is found more often in younger than in older people.
Older people are more often infected by Legionella.
Prevention is by not smoking and avoiding other lung irritants. Frequent hand washing may also be protective. Treatment of acute bronchitis typically involves rest, paracetamol (acetaminophen), and NSAIDs to help with the fever. Cough medicine has little support for its use and is not recommended in children less than six years of age. There is tentative evidence that salbutamol may be useful in those with wheezing; however, it may result in nervousness and tremors. Antibiotics should generally not be used. An exception is when acute bronchitis is due to pertussis. Tentative evidence supports honey and pelargonium to help with symptoms. Getting plenty of rest and fluids is also often recommended.
Post-viral cough can be resistant to treatment. Post-viral cough usually goes away on its own; however, cough suppressants containing codeine may be prescribed. A study has claimed theobromine in dark chocolate is more effective.
There is low or very-low quality evidence that probiotics may be better than placebo in preventing acute URTIs. Vaccination against influenza viruses, adenoviruses, measles, rubella, "Streptococcus pneumoniae", "Haemophilus influenzae", diphtheria, "Bacillus anthracis", and "Bordetella pertussis" may prevent them from infecting the URT or reduce the severity of the infection.
In June 2009, the United States Department of Agriculture (USDA) Animal and Plant Health Inspection Service (APHIS) approved the first canine influenza vaccine. This vaccine must be given twice initially with a two-week break, then annually thereafter.
The Centers for Disease Control describe protocol for treating sinusitis while at the same time discouraging overuse of antibiotics:
- Target likely organisms with first-line drugs: Amoxicillin, Amoxicillin/Clavulanate
- Use shortest effective course: Should see improvement in 2–3 days. Continue treatment for 7 days after symptoms improve or resolve (usually a 10–14 day course).
- Consider imaging studies in recurrent or unclear cases: some sinus involvement is frequent early in the course of uncomplicated viral URI
Treatment comprises symptomatic support usually via analgesics for headache, sore throat and muscle aches. Moderate exercise in sedentary subjects with naturally acquired URTI probably does not alter the overall severity and duration of the illness. No randomized trials have been conducted to ascertain benefits of increasing fluid intake.
The presence of an upper respiratory tract infection in a dog that has been vaccinated for the other major causes of kennel cough increases suspicion of infection with canine influenza, especially in areas where the disease has been documented. A serum sample from a dog suspected of having canine influenza can be submitted to a laboratory that performs PCR tests for this virus.
Prevention of bronchiolitis relies strongly on measures to reduce the spread of the viruses that cause respiratory infections (that is, handwashing, and avoiding exposure to those symptomatic with respiratory infections). In addition to good hygiene an improved immune system is a great tool for prevention. One way to improve the immune system is to feed the infant with breast milk, especially during the first month of life. Immunizations are available for premature infants who meet certain criteria (some cardiac and respiratory disorders) such as Palivizumab (a monoclonal antibody against RSV). Passive immunization therapy requires monthly injections during winter.
The distinction between viral upper respiratory tract infections is loosely based on the location of symptoms with the common cold affecting primarily the nose, pharyngitis the throat, and bronchitis the lungs. However, there can be significant overlap and multiple areas can be affected. The common cold is frequently defined as nasal inflammation with varying amount of throat inflammation. Self-diagnosis is frequent. Isolation of the viral agent involved is rarely performed, and it is generally not possible to identify the virus type through symptoms.
Flavorings-related lung disease can be prevented with the use of engineering controls (e.g. exhaust hoods or closed systems), personal protective equipment, monitoring of potentially affected workers, worker education, and by not using lung-disease-causing flavorings.