Swallowing, Speech and Language — The Crossing Where the Food Path Meets the Airway

Before a Single Spoonful

By the hospital bedside of a father admitted for a stroke, one of the things a caregiver hears early on is, unexpectedly, talk about eating. Meals will come after checking his ability to swallow, they say. Limbs were the worry, but eating is what has stopped first.

The speech is different, too. The father's intention to speak is clear, but his articulation is slurred and slow, making it hard to understand. Another patient in the same room has clear articulation, but the words they are looking for do not come out, or the wrong words pop out.

Eating and speaking use the same mouth and throat. Yet when they break down, they do so in different places and in different ways. Eating is a path running from the mouth down to the stomach, and that path meets the airway in the throat once. Speaking has two layers. There are muscles that shape sound, and ahead of that, a brain that decides which words to choose.

Swallowing — From Mouth to Stomach, A Single Intersection

Swallowing a single bite requires more than 30 nerves and muscles to move together. Among them, voluntary actions and automatic actions are mixed together (PMID 18940636). This chain has two jobs to accomplish: sending food from the mouth to the stomach, and protecting the airway all the while (PMID 18940636).

Swallowing is divided into three stages depending on where the food bolus currently is: the oral stage, the pharyngeal stage, and the esophageal stage (PMID 18940636). The way food moves from inside the mouth toward the throat differs between chewing food and drinking water (PMID 18940636). This means that even in a single swallow, the movement changes depending on what is being swallowed.

The pharynx is the space inside the throat extending downward from behind the nose and mouth, serving as a shared passageway for food and air. At the lower end of the pharynx, the path splits into two: the front is the airway leading to the lungs, and the back is the esophagus leading to the stomach. The larynx is the entrance to the airway and houses the vocal cords that produce sound. Food passes directly behind this airway entrance to enter the esophagus. Therefore, inside the throat, the food path and the airway cross each other. This article calls that spot the intersection.

At the intersection, the road must be blocked when a train passes. Inside the throat, the moment food passes through the pharynx, the larynx is pulled upward and forward, closing the laryngeal entrance. This closure is multi-layered. The airway closes at multiple levels, from the vocal cords below to the epiglottis above.

The epiglottis is a leaf-shaped flap standing above the laryngeal entrance behind the base of the tongue. An anatomical study that sliced and examined 25 human larynges thinly has provided an explanation of how the laryngeal entrance closes. This study views the entrance by dividing it into front and back. The front is closed along with the downward movement of the epiglottis, and these two movements rely on the pressure applied to the fatty tissue in front of the flap. The back is closed along with the closure of the vocal cords below it, and these two movements rely on the small cartilages attached to the rear ends of the vocal cords pulling inward (PMID 9438107).

The brain conducts this short chain. Each and every swallow is coordinated by a network of nerve cells spanning the deep lower part of the brain, the cerebral cortex covering the surface, and the cerebellum in the lower back (PMID 38517302). Therefore, whether this network is injured, or the nerves leading from it to the muscles or the muscles themselves are injured, swallowing can be shaken (PMID 38517302). Swallowing is a chain that accomplishes sending food and blocking the airway at once, and if the timing of the two tasks misaligns or a gap forms on the blocking side, food can leak into the airway. There are also ways it leaks after swallowing. If the pushing force is weak or the esophageal entrance opens insufficiently, food remains in the throat, and that food can enter when the airway reopens.

Leaking at the Crossing — When the Alarm Sounds and When It Does Not

Dysphagia is a condition where swallowing does not work properly or is out of sync. It can occur when the movement or structure of the mouth, pharynx, larynx, or esophagus is damaged in any of those areas (PMID 18940636). It appears in various diseases, and the most common and most studied is dysphagia following a stroke (PMID 38517302).

Aspiration is when food, water, or saliva enters the airway below the vocal cords. Coughing and spluttering is the image of aspiration that commonly comes to mind. Common signs that indicate food has entered the airway side are coughing, throat-clearing, and changes in voice (PMID 38301043). Among them, coughing is the alarm of the crossing. It is an action that signals something has entered the airway side and pushes it back out.

There are also times when the alarm does not sound. Silent aspiration is aspiration that occurs without outward signs such as coughing, throat-clearing, or voice changes. It is also called asymptomatic aspiration. Two sources answer how common silent aspiration is, looking at different groups of patients.

The first source is a small study. The acute phase is the initial period right after a stroke has occurred. This study examined 64 patients with suspected dysphagia among severe stroke patients who did not improve even after the acute phase passed, using imaging tests. 66% aspirated, and half of those who aspirated had silent aspiration (PMID 16487410). In the same study, bedside examination findings failed to distinguish whether aspiration was present (PMID 16487410).

The second source is a paper focusing on head and neck surgery patients that summarizes values from prior studies. It notes that 20 to 30% of people who aspirate do so silently, and bedside swallowing exams can miss up to half of silent aspirations (PMID 38301043). These studies mix patients with strokes, brain injuries, head and neck cancer treatments, and long-term tracheostomy. Therefore, these values cannot be read as the rate solely for stroke patients (PMID 38301043). The "half" from the first source and the percentages from the second source are values from different populations.

Therefore, one conclusion comes together from both sources. We cannot say that food has not entered the airway simply because no cough occurred. Pneumonia and repeated respiratory infections are things mentioned alongside silent aspiration. It can also appear as an unnoticed decrease in the amount eaten and weight loss (PMID 38301043).

Crossings where the alarm does not sound can also be looked into. There are instrumental exams that directly view swallowing behavior using an endoscope or imaging, and silent aspiration is revealed in these exams (PMID 38301043). The trend drawn by recent literature has three steps. Validated simple screening first filters out people who need a closer look, followed by a swallowing evaluation directly conducted by medical staff if needed, and then proceeding to a videofluoroscopic swallowing study. It is considered that accurate diagnosis and treatment plans emerge only after going through this flow (PMID 42496759). What the tests look at and how they are read is covered in "How Do We Know Swallowing Has Failed". Even when there is no alarm, something can pass through, and there are ways to see it.

Swallowing After Stroke — How Common It Is, and What It Moves With

Just after a stroke, roughly four out of ten patients in the acute phase experience dysphagia. Recalculating by pooling data from 42 studies involving 26,366 people, 42% had dysphagia (PMID 35562660).

This number varies depending on the tool used to find it. In a 2005 systematic review pooling 24 studies, when filtered out by a simple screen, the rate of dysphagia was 37% to 45%. When found by a clinical examination where clinicians examine the patient directly, it was 51% to 55%, and when found by instrumental tests, it was 64% to 78% (PMID 16269630). In the aforementioned analysis of 42% as well, studies that used instrumental tests showed higher rates (PMID 35562660). The more carefully you look, the more you find. Therefore, the answer to "how common is it?" must be read alongside how carefully it was looked for.

Dysphagia moves together with pneumonia. Stroke patients with dysphagia developed pneumonia more frequently, and the difference was even greater in patients where aspiration was confirmed (PMID 16269630). In the previous analysis pooling acute-phase patients as well, both pneumonia and mortality were higher on the side with dysphagia (PMID 35562660). Post-stroke pneumonia occurs in about 12% of acute stroke patients and is the most common infectious complication following a stroke (PMID 42095755).

What these numbers show is an association in which dysphagia and pneumonia appear together. The mere fact that they appear together does not mean one causes the other. A review covering aspiration and infection in older adults views dysphagia and aspiration as important factors in pneumonia. However, it considers that pneumonia does not occur from those alone without other conditions. Those other conditions are things like disease-causing bacteria settling inside the mouth or a decline in the body's defense capabilities (PMID 15733019). How far one can speak of causality between aspiration and pneumonia is addressed in "Is Aspiration Pneumonia Really Caused by Aspiration?".

Guidelines also move based on this association. The 2026 European Stroke Organisation guidelines on post-stroke pneumonia view two things as already established parts of care to prevent post-stroke pneumonia: screening to identify dysphagia and swallowing management. The specific details of that are covered in separate guidelines (PMID 42095755). The same guidelines do not recommend prophylactic antibiotics used in advance before pneumonia develops. This is because there was no benefit in improving the occurrence or course of pneumonia. And the guidelines themselves disclose that most of their guideline sentences rely on expert consensus rather than high-level evidence (PMID 42095755).

Post-stroke dysphagia is not viewed as a fixed condition once it is determined. Dysphagia in the acute phase is common, but most of it improves. This is what a 2006 study cited earlier noted as background explanation (PMID 16487410). However, the speed and degree of improvement vary from person to person. Therefore, recent literature views post-stroke dysphagia not as a temporary complication that comes in the acute phase, but as a condition that changes over time and requires long-term management (PMID 42496759). The same literature points out things that happen because management systems are disconnected. These are things like failing to be re-evaluated and being kept from eating by mouth longer than necessary (PMID 42496759).

We also looked at who improves better. The older the age and the more severe the stroke, the less swallowing improved. The more severe the initial dysphagia and the presence of aspiration, the less it improved. Those who entered treatment earlier improved more (PMID 37190927). Because these results pool studies that observed patients, one cannot even go so far as to say these factors are the causes. Post-stroke dysphagia is common and moves together with pneumonia, but because it is a changing condition, looking into it again is part of treatment.

Speech and Language Are Different — Slurred Sounds and Missing Words

Let us return to the two people from the opening scene. The father knows what he wants to say, but his sounds are slurred. The other patient has clear sounds, but their words are misaligned. The former is a problem of speech, and the latter is a problem of language. In this article, speech is the bodily movement of shaping sounds and the brain's control over those movements, while language is the brain's work of building and understanding words and sentences. Both can break down when the brain is injured. The problem of the movements that create pronunciation can be separated from the difficulty with language, and making that separation is useful for diagnosis (PMID 35078611). Both can also come together in the same person.

Dysarthria is a condition where the movements of the muscles that make speech sounds become weak, inaccurate, imprecise, or uncoordinated after an injury to the brain or nerves, making speech hard to understand. It can arise from brain injuries like a stroke, or from diseases of the peripheral nerves or muscles. The voice comes from the vocal cords in the larynx, and the lips and tongue shape that sound into individual speech sounds. Speech sounds are made right at the same crossroads where swallowing passes through. The impact of dysarthria does not end with a failure to communicate; it extends to a person's emotional and social life (PMID 28121021).

Aphasia is a state where the ability to produce or understand language breaks down due to brain injury. In this definition, language means words and sentences, not pronunciation. Therefore, assessing aphasia looks at naming, grammar, comprehension and repetition, reading and writing, and even everyday conversation (PMID 35078598). The difficulty of not being able to find the desired word is a common symptom in aphasia (PMID 14872396).

Roughly one-third of people who experience a stroke experience aphasia (PMID 27245310). The extent to which language improves is greatest in the first few weeks after a stroke. Smaller improvements can continue even after that over several years (PMID 30698070). While many people improve significantly, many others are left with lasting language difficulties (PMID 25888529).

Aphasia is an injury to language, not the same thing as the complete loss of the power to think. A brief review notes that even patients who cannot name objects often know what they are trying to point to (PMID 16623057). Even when aphasia is present, non-language cognition such as memory or attention can be assessed with a fair degree of reliability (PMID 30698070). Knowing how severe the aphasia was initially and how intact non-language cognition remains helps the treatment team set their direction. It also helps adjust expectations for how much recovery may happen in the future (PMID 30698070). Aphasia weighs heavily on a person's emotional and social life (PMID 35078598).

Sudden slurring of speech or sudden difficulty with words can be a sign of an urgent illness like a stroke (PMID 35078611). In such cases, time changes the outcome. Whether it is a stroke is confirmed by brain imaging tests at the hospital. Those signs and what to do at that time are covered in What a Stroke Actually Is. Slurred pronunciation and missing words are different breakdowns, and even if speech is blocked, thought has not vanished.

What Treatment Does — The Power to Swallow and the Path of Communication

Treatment for dysphagia splits into two paths. The first path is preventing complications like pneumonia and malnutrition. Adjusting the food eaten, keeping the inside of the mouth clean, and feeding through a tube if needed fall under this, and it remains the cornerstone of treatment today (PMID 38508837). The second path is trying to build up the swallowing strength itself. Various behavioral training methods have shown potential in small studies, and methods that change throat sensation and stimulate the brain or neck are being studied (PMID 38508837). The decision to change the consistency of food is covered in "The Decision to Change Viscosity", and the decision to feed through a tube in "When Eating Is Not Possible — Facing the Tube-Feeding Decision".

The evidence for the second path is most heavily gathered in a Cochrane review. A Cochrane review is a literature synthesis that gathers and analyzes studies by a set procedure, and when new studies accumulate, it updates and republishes the framework. When looking at treatment effects, it mainly gathers trials that randomly divided people to receive treatment and compared them. The 2026 edition gathered 181 random trials with 11,500 people covering patients with dysphagia within six months after a stroke. It then compared various treatments with usual care (PMID 42573047).

To read those results, one yardstick is needed. The certainty of evidence is a grade rating how much we can trust the gathered results as high, moderate, low, or very low. The lower the certainty, the greater the room for conclusions to change when more studies accumulate. Behavioral therapies like swallowing exercises or learning safe swallowing tips tended to reduce swallowing problems more than usual care, but the certainty was very low. Results that chest infections or pneumonia could be reduced had low certainty (PMID 42573047). This abstract does not specify what goes into behavioral therapy.

The method of applying electrical stimulation to the neck and the method of delivering weak current to the brain from outside the head appeared to reduce swallowing problems. The certainty of this result was moderate. However, other results of the same treatments had lower certainty (PMID 42573047). Other treatments like acupuncture, medication, and magnetic stimulation were also compared together. The authors list three things blocking a conclusion and note that large, high-quality trials are needed. They are small trials with a high or unclear risk of skewed results, publication bias where positive results are published more easily, and conflicting results among studies (PMID 42573047).

For exercises training the swallowing muscles of the tongue, jaw, and neck, there is a separately gathered analysis. The authors of the analysis, which re-calculated by gathering 13 studies, conclude that these trainings benefit swallowing function and reduce the risk of food entering the airway side (PMID 42014247). This abstract has no certainty grade. In the preceding Cochrane review as well, behavioral therapy tended to reduce the degree of aspiration measured by imaging tests, and the certainty was low (PMID 42573047).

Guidelines also pick the exercise side. A summary of the stroke guideline issued in 2024 by the US Department of Veterans Affairs and Department of Defense extracted areas important to primary care physicians. Among them, in the area grouping swallowing, aphasia, and cognition, two exercises are picked as having the strongest evidence. They are jaw-pulling exercises against resistance and strength training for breathing muscles (PMID 39832369). However, this guideline differs from the previous Cochrane review in the gathered studies and the analyzed outcomes, so the two evaluations do not share the same yardstick. For any exercise, how frequently, how intensely, and how long it should be done has not been established yet. This is because reports vary back and forth and differ greatly across studies (PMID 36471047). The relationship between the amount of practice and its effect is covered in another piece in this series, "Neuroplasticity".

Beyond swallowing, there are other things that get in the way of keeping treatment going steadily. A synthesis gathering 5 interview studies on the experiences of patients experiencing dysphagia suggests that participating steadily in treatment requires addressing emotional and social life difficulties together. It suggests that support from family, hospitals, and the local community must also be addressed together (PMID 41173268).

Moving on to the language side. Speech therapy is a formal therapy that aims to raise language and communication abilities, and through that, bring them up to the level of conversing in daily life and participating in society. The goal of aphasia therapy is ultimately to improve everyday communication and raise the strength to live day by day (PMID 35076868). Approaches dealing with word-finding difficulty are twofold. One is to revive the brain pathways that recall words. The other is to use other cognitive abilities or remaining abilities to bypass the blocked spot (PMID 14872396).

Augmentative and alternative communication means communication tools that help or replace speech, such as picture boards, letter boards, gestures, and devices. It is a representative way of bypassing a blocked spot. The evidence that this helps adults with severe aphasia came from controlled environments like therapy rooms, and the communication improvements gained that way usually did not translate into daily life (PMID 15513721). This means that transferring communication learned in the therapy room to life outside the therapy room remains a remaining task.

The latest edition of the Cochrane review gathering the effects of speech therapy is the 2016 edition. It gathered 57 random trials with 3,002 people up to September 2015, and looked at comparisons divided into three types (PMID 27245310).

The first is a comparison weighing people who received speech therapy against those who did not, totaling 27 trials and 1,620 people. Speech therapy improved everyday communication, reading, writing, and spoken language in ways that were both clinically and statistically meaningful (PMID 27245310). Statistically meaningful means it is hard to view as a coincidence, and clinically meaningful means it is big enough to feel in life. However, in follow-up measurements taken some time after treatment ended, the number of people was smaller and the benefit was not distinct (PMID 27245310).

The second is a comparison weighing speech therapy against a side given only social support and conversation stimulation, totaling 9 trials and 447 people. Here, there was no evidence of a difference in everyday communication. People who quit midway were more numerous on the social support side (PMID 27245310).

The third is a comparison weighing different speech therapies against each other, totaling 38 trials and 1,242 people. Everyday communication was better on the side with high intensity, large total amount, or longer duration. However, on the side with high intensity or large total amount, the number of people quitting midway was also distinctly higher, which blunted the benefit (PMID 27245310).

The trials were mostly small, and age, timing after stroke, and severity varied wildly (PMID 27245310). The authors' conclusions are as follows. Compared to the untreated side, there is evidence that speech therapy improves everyday communication, reading, writing, and spoken language. There are also some signs that treatments with increased intensity, total amount, or prolonged duration can be beneficial. However, not everyone may be able to accept such treatments (PMID 27245310). This abstract did not attach certainty grades to each result.

There is also intensive speech therapy that strongly pushes people to use speech. According to a study re-analyzing six literature syntheses covering this therapy, while various language and communication indicators improved with this therapy, they were often not better than comparison treatments. Because the effects were similar to other intensive treatments, the authors suggest the core of the effect may be training intensity (PMID 36668725).

The European Stroke Organisation's 2025 aphasia guideline, leaning on low-certainty evidence, recommends increasing the total overall amount of speech therapy to over 20 hours. It also suggests moving toward higher intensity and frequency (PMID 40401776). The same guideline suggests personalized speech therapy for an individual as well as digital and group-format therapies. Regarding the method of delivering weak current to the brain, it issues an expert consensus that the evidence is very weak and it should be used only within high-quality trials (PMID 40401776). On the side where intensity and total amount were increased, the number of dropouts also increased. This was reported by the earlier Cochrane review, and the guideline abstract does not note this point separately.

What cuts deepest in deciding how much someone will improve is how severe the aphasia was initially. Even so, it is difficult to predict in advance how far an individual will improve (PMID 30698070). Age, sex, and education level were not confirmed as reliable predictive factors (PMID 25888529). How effective treatment is when done from right after a stroke to several months later is not yet clear (PMID 30698070). The timing of improvement and how much treatment adds to it are covered in another piece in this series, "Neuroplasticity".

The evidence on the dysarthria side is thinner. A 2017 Cochrane review covering dysarthria after stroke or other brain injury gathered 5 small trials with 234 people. Among them, there were no trials comparing against an untreated side. They were trials compared with attention-only comparison groups or usual care (PMID 28121021). There was no evidence of effects lasting from three to nine months after treatment ended. There was a little evidence that speech sound itself improved right after treatment, but the certainty was very low (PMID 28121021). While stating that benefits and risks are not yet known, the authors write that people with dysarthria after a stroke or brain injury should continue receiving treatment according to clinical guidelines (PMID 28121021). That evidence to confirm effectiveness is still lacking is a different statement from confirming that there is no effect.

Treatments for swallowing and language are accumulating evidence pointing toward improvement, but many areas have low certainty, and the goal is everyday meals and communication.

Holding It Up to One Person

Let's hold the preceding facts up to our initial father, one by one. In situations where a meal does not arrive immediately, there is a sequence that checks swallowing first. It is a sequence that screens first with a simple check, and if necessary, leads to a bedside swallowing evaluation and a videofluoroscopic swallowing study conducted directly by medical staff (PMID 42496759). Who undergoes which tests is covered in "How Do We Know Swallowing Has Failed".

The mere fact that he did not cough during a meal does not tell us whether the crosswalk is safe. Findings from a bedside examination alone make it difficult to screen for aspiration, which is often revealed only by a test that directly visualizes the act of swallowing on video (PMID 16487410). However, that study examined 64 severe patients with suspected swallowing who did not improve even after the acute phase had passed. Therefore, the percentages derived from there cannot be transferred to a father who was just admitted. The one thing that can be transferred is the point that "the absence of cough alone does not mean it is safe." Nor does it end with a single test result. Because swallowing disorder is a state that changes over time, being reevaluated is part of management, and a period of not being able to eat by mouth that becomes longer than necessary is also a separate issue to examine (PMID 42496759).

When another patient in the same room looks at a cup and says a different word, he may not actually fail to know what a cup is. Patients who cannot name things often still know what they are trying to point to (PMID 16623057). How intact cognition outside of language is can also be evaluated separately to some extent (PMID 30698070). There are also alternative paths of communication, such as communication boards or gestures. Making those paths usable outside the therapy room remains a task yet to be accomplished (PMID 15513721).

Where to Look Next

• If you are curious about the tests that determine whether swallowing has failed, continue to "How Do We Know Swallowing Has Failed" from the series "Dysphagia — Between Eating and Breathing".
• The causal link between aspiration and pneumonia is examined in "Is Aspiration Pneumonia Really Caused by Aspiration?" from the same series.
• Changing the consistency of food and tube feeding are covered in "The Decision to Change Viscosity" and "When Eating Is Not Possible — Facing the Tube-Feeding Decision" from the same series.
• The principles by which an injured brain forms new connections continue in "How the Central Nervous System Recovers — A Primer on Neuroplasticity", and the roots of therapies that strongly encourage the use of the limb continue in "Able But Unused — An Introduction to Learned Non-Use and CIMT".
• Emergency signs of stroke are covered in "What a Stroke Actually Is — Blocked, Burst, and the Clock".
• The timeline of recovery and the amount of practice are covered in another piece in this series, "Neuroplasticity", while what the various therapists you meet in the hospital do is covered in another piece in this series, "Rehabilitation".

Source Papers

• "Anatomy and physiology of feeding and swallowing: normal and abnormal." (2008) — PMID 18940636 — Over 30 nerves and muscles, the dual tasks of food transport and airway protection, three stages of the mouth, pharynx, and esophagus, and dysphagia caused by abnormalities anywhere in the mouth, pharynx, larynx, or esophagus.
• "Anatomical considerations of closure of the laryngeal vestibule during swallowing." (1997) — PMID 9438107 — Dissection of 25 normal adult larynges, showing that anterior closure of the laryngeal inlet is related to epiglottic descent and posterior closure to vocal fold closure, both relying on arytenoid cartilage adduction.
• "The neurorehabilitation of post-stroke dysphagia: Physiology and pathophysiology." (2025) — PMID 38517302 — Neural networks in the brainstem, cerebral cortex, and cerebellum coordinate swallowing; definition of dysphagia; post-stroke dysphagia is the most common and most widely studied.
• "Oropharyngeal dysphagia after the acute phase of stroke: predictors of aspiration." (2006) — PMID 16487410 — 64 patients with severe stroke and suspected dysphagia unrecovered after the acute phase; 66% showed aspiration on instrumental examination, half of which was silent; aspiration cannot be predicted by clinical examination; acute dysphagia is common but mostly improves.
• "The risk factors for silent aspiration: A retrospective case series and literature review." (2024) — PMID 38301043 — Summary of prior studies in head and neck surgery-focused literature: combining multiple patient groups, 20 to 30% of patients with aspiration have silent aspiration; clinical examination misses up to 50%; cough, throat-clearing, and voice changes are signals; results in pneumonia and weight loss; high-risk groups prioritize instrumental examination.
• "Post-Stroke Dysphagia: Moving Toward Evidence-Based, Recovery-Oriented Systems of Care." (2026) — PMID 42496759 — Screening → clinical evaluation → swallowing imaging; dysphagia is a changing state requiring long-term management; fragmented systems lead to missed reevaluations and unnecessarily prolonged fasting.
• "Prevalence of dysphagia and risk of pneumonia and mortality in acute stroke patients: a meta-analysis." (2022) — PMID 35562660 — 42 studies and 26,366 patients, acute phase prevalence of 42%, odds ratio for pneumonia and mortality approximately 4, higher prevalence found on instrumental examination.
• "Dysphagia after stroke: incidence, diagnosis, and pulmonary complications." (2005) — PMID 16269630 — 24 studies, screening 37–45%, clinical 51–55%, and instrumental 64–78%; increased risk of pneumonia with dysphagia and aspiration.
• "European stroke organisation guideline on stroke-associated pneumonia." (2026) — PMID 42095755 — Pneumonia after stroke occurs in about 12% and is the most common infectious complication; screening and swallowing management are established elements of prevention; prophylactic antibiotics are not recommended; mostly expert consensus.
• "Aspiration and infection in the elderly : epidemiology, diagnosis and management." (2005) — PMID 15733019 — Dysphagia and aspiration are important factors but are not sufficient for aspiration pneumonia without other risk factors.
• "Demographic and clinical factors associated with recovery of poststroke dysphagia: A meta-analysis." (2023) — PMID 37190927 — 28 observational studies; older age, severe stroke, severe dysphagia at admission, and aspiration are unfavorable for recovery; early intervention is favorable.
• "Types of motor speech impairments associated with neurologic diseases." (2022) — PMID 35078611 — Distinguishing motor speech disorders independently of language problems is useful for diagnosis; sudden speech abnormalities can be a sign of acute conditions such as stroke.
• "Interventions for dysarthria due to stroke and other adult-acquired, non-progressive brain injury." (2017) — PMID 28121021 — Definition of dysarthria; 5 small RCTs and 234 patients; no no-treatment comparison; no evidence of sustained effects and limited evidence for immediate impairment-level effects; continue rehabilitation according to guidelines.
• "Assessment of language impairment and function." (2022) — PMID 35078598 — Assessment of aphasia examining naming, grammar, comprehension, repetition, reading, writing, and everyday conversation; psychosocial impact of aphasia.
• "Management of anomia." (2004) — PMID 14872396 — Word-finding difficulty is a common symptom of aphasia; restitutive and compensatory approaches.
• "Naming disorders anomic aphasia." (2005) — PMID 16623057 — Patients with naming impairments clearly understand what they intend to name but cannot recall the word.
• "What is Functional Communication? A Theoretical Framework for Real-World Communication Applied to Aphasia Rehabilitation." (2022) — PMID 35076868 — Aphasia is a disorder of language; the goal of rehabilitation is to improve everyday communication.
• "Language recovery following stroke." (2019) — PMID 30698070 — Language recovery is greatest in the first few weeks but can continue for years; initial severity is the biggest determining factor; individual prediction is difficult; effectiveness of acute and subacute treatment is unclear; cognitive assessment is possible even in aphasia.
• "Dysphagia after stroke: research advances in treatment interventions." (2024) — PMID 38508837 — Protective interventions (dietary modification, oral hygiene, nutrition) and rehabilitative interventions (behavioral strategies, pharmacotherapy, neurostimulation).
• "Swallowing therapy for dysphagia in acute and subacute stroke." (2026) — PMID 42573047 — Cochrane 2026 edition, 181 RCTs, 11,500 patients, compared with usual care; behavioral therapy reduces swallowing problems (very low certainty) and pneumonia (low certainty); electrical stimulation and tDCS have moderate certainty; limitations of small trials and bias.
• "Efficacy of swallowing muscle training for the treatment of patients with post-stroke dysphagia: A systematic review and meta-analysis." (2026) — PMID 42014247 — 13 studies, swallowing muscle training reduces the risk of penetration and aspiration.
• "Stroke Rehabilitation: Synopsis of the 2024 U.S. Department of Veterans Affairs and U.S. Department of Defense Clinical Practice Guidelines." (2025) — PMID 39832369 — Strongest recommendation in the swallowing domain: resistive jaw-pulling and respiratory muscle strength training.
• "Dosages of swallowing exercises in stroke rehabilitation: a systematic review." (2023) — PMID 36471047 — 54 studies, reporting of swallowing exercise dosage is inconsistent and varies widely.
• "Experiences and needs of stroke patients living with dysphagia during rehabilitation: a qualitative meta-synthesis." (2026) — PMID 41173268 — 5 qualitative studies; psychosocial difficulties and gaps in support must be addressed together to improve engagement.
• "Augmentative and Alternative Communication (AAC) for adults with severe aphasia: where we stand and how we can go further." (2004) — PMID 15513721 — AAC supported in controlled treatment environments; generalization to daily life usually does not occur.
• "Speech and language therapy for aphasia following stroke." (2016) — PMID 27245310 — Cochrane 2016 edition (latest), about one-third of people who experience a stroke have aphasia, 57 RCTs, 3,002 patients, improvement in everyday communication, reading, writing, and expressive language compared with no treatment (unclear at follow-up), no evidence of difference compared with social support, higher intensity, higher dosage, and longer treatment are better but increase dropouts, not acceptable to everyone.
• "European Stroke Organisation (ESO) guideline on aphasia rehabilitation." (2025) — PMID 40401776 — Recommends a total volume of 20 hours or more based on low-quality evidence, suggests increasing intensity and frequency, suggests personalized, digital, and group therapy, tDCS only within high-quality trials (expert consensus).
• "Effectiveness of Constraint-Induced Language Therapy for Aphasia: Evidence From Systematic Reviews and Meta-Analyses." (2023) — PMID 36668725 — CILT effectiveness is similar to other intensive treatments; training intensity may be the key.
• "Factors predicting post-stroke aphasia recovery." (2015) — PMID 25888529 — Lesion location and size, aphasia type and severity, and treatment are major factors; age, sex, and education are not robust predictors.