diff --git a/pipeline/outputs/logdna.md b/pipeline/outputs/logdna.md
index ef44125a9..bd9f0912b 100644
--- a/pipeline/outputs/logdna.md
+++ b/pipeline/outputs/logdna.md
@@ -1,3 +1,7 @@
+---
+description: Send logs to LogDNA
+---
+
 # LogDNA
 
 The _LogDNA_ output plugin sends logs and events to a LogDNA-compliant service.
@@ -10,6 +14,7 @@ This plugin uses the following configuration parameters:
 | :--- | :--- | :--- |
 | `api_key` | Required. The API key to get access to the service. | _none_ |
 | `app` | Name of the application. This value is automatically discovered on each record. If no value is found, the default value is used. | `Fluent Bit` |
+| `exclude_promoted_keys` | When enabled, keys promoted to the top-level line object (`meta`, `level`, `severity`, `app`, `file`) are excluded from the `line` body to avoid duplication. | `false` |
 | `file` | Optional name of a file being monitored. This value is only set if the record doesn't contain a reference to it. | _none_ |
 | `hostname` | Name of the local machine or device where Fluent Bit is running. If no value is specified, Fluent Bit will look up the hostname and auto-populate its value. If Fluent Bit is unable to find a value, it will set the value `unknown` instead. | _none_ |
 | `ip` | The IP address of the local hostname. This value is optional. | _none_ |
@@ -24,7 +29,7 @@ This plugin uses the following configuration parameters:
 
 The LogDNA output plugin can automatically discover and enrich records with additional content.
 
-When the plugin processes a record or log, it searches for specific key names that might contain context for the record in question. The following table describe these keys and the discovery logic:
+When the plugin processes a record or log, it searches for specific key names that might contain context for the record in question. The following table describes these keys and the discovery logic:
 
 | Key | Description |
 | :--- | :--- |
@@ -33,6 +38,8 @@ When the plugin processes a record or log, it searches for specific key names th
 | `app` | If the record contains a key called `app`, it will populate the context `app` with the value found, otherwise it will use the value set for `app` in the configuration property. |
 | `meta` | If the record contains a key called `meta`, it will populate the context `meta` with the value found. |
 
+These keys are referred to as _promoted keys_ because their values are lifted to the top-level line object sent to LogDNA. By default, promoted keys also remain present in the `line` JSON body. To remove them from the `line` body and avoid duplication, set `exclude_promoted_keys` to `true`.
+
 ## Example configuration
 
 The following example configuration uses a `dummy` input and `logdna` output:
@@ -66,22 +73,22 @@ pipeline:
 
 ```text
 [SERVICE]
-    Flush     1
-    Log_Level info
+  Flush     1
+  Log_Level info
 
 [INPUT]
-    Name      dummy
-    Dummy     {"log":"a simple log message", "severity": "INFO", "meta": {"s1": 12345, "s2": true}, "app": "Fluent Bit"}
-    Samples   1
+  Name      dummy
+  Dummy     {"log":"a simple log message", "severity": "INFO", "meta": {"s1": 12345, "s2": true}, "app": "Fluent Bit"}
+  Samples   1
 
 [OUTPUT]
-    Name      logdna
-    Match     *
-    Api_Key   YOUR_API_KEY_HERE
-    Hostname  my-hostname
-    Ip        192.168.1.2
-    Mac       aa:bb:cc:dd:ee:ff
-    Tags      aa, bb
+  Name      logdna
+  Match     *
+  Api_Key   YOUR_API_KEY_HERE
+  Hostname  my-hostname
+  Ip        192.168.1.2
+  Mac       aa:bb:cc:dd:ee:ff
+  Tags      aa, bb
 ```
 
 {% endtab %}
@@ -116,3 +123,61 @@ Your record will be available and visible in your LogDNA dashboard after a few s
 In your LogDNA dashboard, go to the top filters and mark the Tags `aa` and `bb`, then you will be able to see your records as shown:
 
 ![LogDNA dashboard](../../.gitbook/assets/logdna.png)
+
+## Exclude promoted keys
+
+When a record contains promoted keys (`level`, `severity`, `app`, `file`, or `meta`), those values are lifted to the top-level line object. By default, they also appear inside the `line` JSON body, which can result in duplicate fields.
+
+Setting `exclude_promoted_keys` to `true` removes the promoted keys from the `line` body, keeping each field in only one place.
+
+### Example
+
+Given this input record:
+
+```json
+{"log": "starting service", "severity": "INFO", "app": "my-service", "host": "web-01"}
+```
+
+With the default behavior (`exclude_promoted_keys: false`), the `line` body sent to LogDNA contains all fields:
+
+```json
+{"log": "starting service", "severity": "INFO", "app": "my-service", "host": "web-01"}
+```
+
+With `exclude_promoted_keys: true`, the `line` body contains only the non-promoted fields:
+
+```json
+{"log": "starting service", "host": "web-01"}
+```
+
+The `severity` and `app` values are still sent, but only as top-level fields in the LogDNA line object, not duplicated inside `line`.
+
+### Configuration
+
+{% tabs %}
+{% tab title="fluent-bit.yaml" %}
+
+```yaml
+pipeline:
+  outputs:
+    - name: logdna
+      match: '*'
+      api_key: YOUR_API_KEY_HERE
+      hostname: my-hostname
+      exclude_promoted_keys: true
+```
+
+{% endtab %}
+{% tab title="fluent-bit.conf" %}
+
+```text
+[OUTPUT]
+  Name                  logdna
+  Match                 *
+  Api_Key               YOUR_API_KEY_HERE
+  Hostname              my-hostname
+  Exclude_Promoted_Keys true
+```
+
+{% endtab %}
+{% endtabs %}